Search Results for author: Yu-Gang Jiang

Found 218 papers, 110 papers with code

Hierarchical Visual-Textual Graph for Temporal Activity Localization via Language

1 code implementation ECCV 2020 Shaoxiang Chen, Yu-Gang Jiang

Temporal Activity Localization via Language (TALL) in video is a recently proposed challenging vision task, and tackling it requires fine-grained understanding of the video content, however, this is overlooked by most of the existing works.

Sentence

BrokenVideos: A Benchmark Dataset for Fine-Grained Artifact Localization in AI-Generated Videos

no code implementations25 Jun 2025 Jiahao Lin, Weixuan Peng, Bojia Zi, Yifeng Gao, Xianbiao Qi, Xingjun Ma, Yu-Gang Jiang

Through extensive evaluation, we demonstrate that BrokenVideos establishes a critical foundation for benchmarking and advancing research on artifact localization in generative video models.

Artifact Detection Benchmarking +1

NAP-Tuning: Neural Augmented Prompt Tuning for Adversarially Robust Vision-Language Models

no code implementations15 Jun 2025 Jiaming Zhang, Xin Wang, Xingjun Ma, Lingyu Qiu, Yu-Gang Jiang, Jitao Sang

Vision-Language Models (VLMs) such as CLIP have demonstrated remarkable capabilities in understanding relationships between visual and textual data through joint embedding spaces.

Adversarial Robustness

GenBreak: Red Teaming Text-to-Image Generators Using Large Language Models

no code implementations11 Jun 2025 Zilong Wang, Xiang Zheng, Xiaosen Wang, Bo wang, Xingjun Ma, Yu-Gang Jiang

Recent research on red-teaming and adversarial attacks against T2I models has notable limitations: some studies successfully generate highly toxic images but use adversarial prompts that are easily detected and blocked by safety filters, while others focus on bypassing safety mechanisms but fail to produce genuinely harmful outputs, neglecting the discovery of truly high-risk prompts.

Large Language Model Red Teaming

ControlThinker: Unveiling Latent Semantics for Controllable Image Generation through Visual Reasoning

1 code implementation4 Jun 2025 Feng Han, Yang Jiao, Shaoxiang Chen, Junhao Xu, Jingjing Chen, Yu-Gang Jiang

The field of controllable image generation has seen significant advancements, with various architectures improving generation layout consistency with control signals.

Image Generation Visual Reasoning

CreatiDesign: A Unified Multi-Conditional Diffusion Transformer for Creative Graphic Design

no code implementations25 May 2025 HUI ZHANG, Dexiang Hong, Maoke Yang, Yutao Chen, Zhao Zhang, Jie Shao, Xinglong Wu, Zuxuan Wu, Yu-Gang Jiang

Graphic design plays a vital role in visual communication across advertising, marketing, and multimedia entertainment.

Marketing

OmniGenBench: A Benchmark for Omnipotent Multimodal Generation across 50+ Tasks

1 code implementation24 May 2025 Jiayu Wang, Yang Jiao, Yue Yu, Tianwen Qian, Shaoxiang Chen, Jingjing Chen, Yu-Gang Jiang

Recent breakthroughs in large multimodal models (LMMs), such as the impressive GPT-4o-Native, have demonstrated remarkable proficiency in following general-purpose instructions for image generation.

Image Generation Instruction Following +1

SAMA: Towards Multi-Turn Referential Grounded Video Chat with Large Language Models

no code implementations24 May 2025 Ye Sun, Hao Zhang, Henghui Ding, Tiehua Zhang, Xingjun Ma, Yu-Gang Jiang

Achieving fine-grained spatio-temporal understanding in videos remains a major challenge for current Video Large Multimodal Models (Video LMMs).

Benchmarking Video Grounding

SafeVid: Toward Safety Aligned Video Large Multimodal Models

no code implementations17 May 2025 Yixu Wang, Jiaxin Song, Yifeng Gao, Xin Wang, Yang Yao, Yan Teng, Xingjun Ma, Yingchun Wang, Yu-Gang Jiang

SafeVid uniquely transfers robust textual safety alignment capabilities to the video domain by employing detailed textual video descriptions as an interpretive bridge, facilitating LLM-based rule-driven safety reasoning.

Safety Alignment

WorldPM: Scaling Human Preference Modeling

1 code implementation15 May 2025 Binghai Wang, Runji Lin, Keming Lu, Le Yu, Zhenru Zhang, Fei Huang, Chujie Zheng, Kai Dang, Yang Fan, Xingzhang Ren, An Yang, Binyuan Hui, Dayiheng Liu, Tao Gui, Qi Zhang, Xuanjing Huang, Yu-Gang Jiang, Bowen Yu, Jingren Zhou, Junyang Lin

Motivated by scaling laws in language modeling that demonstrate how test loss scales as a power law with model and dataset sizes, we find that similar laws exist in preference modeling.

Language Modeling Language Modelling

A Comprehensive Survey in LLM(-Agent) Full Stack Safety: Data, Training and Deployment

no code implementations22 Apr 2025 Kun Wang, Guibin Zhang, Zhenhong Zhou, Jiahao Wu, Miao Yu, Shiqian Zhao, Chenlong Yin, Jinhu Fu, Yibo Yan, Hanjun Luo, Liang Lin, Zhihao Xu, Haolang Lu, Xinye Cao, Xinyun Zhou, Weifei Jin, Fanci Meng, Shicheng Xu, Junyuan Mao, Yu Wang, Hao Wu, Minghe Wang, Fan Zhang, Junfeng Fang, Wenjie Qu, Yue Liu, Chengwei Liu, Yifan Zhang, Qiankun Li, Chongye Guo, Yalan Qin, Zhaoxin Fan, Kai Wang, Yi Ding, Donghai Hong, Jiaming Ji, Yingxin Lai, Zitong Yu, Xinfeng Li, Yifan Jiang, Yanhui Li, Xinyu Deng, Junlin Wu, Dongxia Wang, Yihao Huang, Yufei Guo, Jen-tse Huang, Qiufeng Wang, Xiaolong Jin, Wenxuan Wang, Dongrui Liu, Yanwei Yue, Wenke Huang, Guancheng Wan, Heng Chang, Tianlin Li, Yi Yu, Chenghao Li, Jiawei Li, Lei Bai, Jie Zhang, Qing Guo, Jingyi Wang, Tianlong Chen, Joey Tianyi Zhou, Xiaojun Jia, Weisong Sun, Cong Wu, Jing Chen, Xuming Hu, Yiming Li, Xiao Wang, Ningyu Zhang, Luu Anh Tuan, Guowen Xu, Jiaheng Zhang, Tianwei Zhang, Xingjun Ma, Jindong Gu, Liang Pang, Xiang Wang, Bo An, Jun Sun, Mohit Bansal, Shirui Pan, Lingjuan Lyu, Yuval Elovici, Bhavya Kailkhura, Yaodong Yang, Hongwei Li, Wenyuan Xu, Yizhou Sun, Wei Wang, Qing Li, Ke Tang, Yu-Gang Jiang, Felix Juefei-Xu, Hui Xiong, XiaoFeng Wang, DaCheng Tao, Philip S. Yu, Qingsong Wen, Yang Liu

Currently, existing surveys on LLM safety primarily focus on specific stages of the LLM lifecycle, e. g., deployment phase or fine-tuning phase, lacking a comprehensive understanding of the entire "lifechain" of LLMs.

Model Editing

SimpleAR: Pushing the Frontier of Autoregressive Visual Generation through Pretraining, SFT, and RL

1 code implementation15 Apr 2025 Junke Wang, Zhi Tian, Xun Wang, Xinyu Zhang, Weilin Huang, Zuxuan Wu, Yu-Gang Jiang

This work presents SimpleAR, a vanilla autoregressive visual generation framework without complex architecure modifications.

Inference Optimization

OmniSVG: A Unified Scalable Vector Graphics Generation Model

no code implementations8 Apr 2025 Yiying Yang, Wei Cheng, Sijin Chen, Xianfang Zeng, Jiaxu Zhang, Liao Wang, Gang Yu, Xingjun Ma, Yu-Gang Jiang

Scalable Vector Graphics (SVG) is an important image format widely adopted in graphic design because of their resolution independence and editability.

model Vector Graphics

UniToken: Harmonizing Multimodal Understanding and Generation through Unified Visual Encoding

1 code implementation6 Apr 2025 Yang Jiao, Haibo Qiu, Zequn Jie, Shaoxiang Chen, Jingjing Chen, Lin Ma, Yu-Gang Jiang

We introduce UniToken, an auto-regressive generation model that encodes visual inputs through a combination of discrete and continuous representations, enabling seamless integration of unified visual understanding and image generation tasks.

Image Generation

CoMP: Continual Multimodal Pre-training for Vision Foundation Models

1 code implementation24 Mar 2025 Yitong Chen, Lingchen Meng, Wujian Peng, Zuxuan Wu, Yu-Gang Jiang

Pre-trained Vision Foundation Models (VFMs) provide strong visual representations for a wide range of applications.

cross-modal alignment

Revealing the Implicit Noise-based Imprint of Generative Models

no code implementations12 Mar 2025 Xinghan Li, Jingjing Chen, Yue Yu, Xue Song, Haijun Shan, Yu-Gang Jiang

Furthermore, we design a new pipeline that pioneers the use of noise patterns, derived from a noise-based imprint extractor, alongside other visual features for AI-generated image detection, resulting in a significant improvement in performance.

AniGaussian: Animatable Gaussian Avatar with Pose-guided Deformation

no code implementations24 Feb 2025 Mengtian Li, Shengxiang Yao, Chen Kai, Zhifeng Xie, Keyu Chen, Yu-Gang Jiang

Recent advancements in Gaussian-based human body reconstruction have achieved notable success in creating animatable avatars.

AnyRefill: A Unified, Data-Efficient Framework for Left-Prompt-Guided Vision Tasks

no code implementations16 Feb 2025 Ming Xie, Chenjie Cao, Yunuo Cai, xiangyang xue, Yu-Gang Jiang

In this paper, we present a novel Left-Prompt-Guided (LPG) paradigm to address a diverse range of reference-based vision tasks.

DuMo: Dual Encoder Modulation Network for Precise Concept Erasure

1 code implementation2 Jan 2025 Feng Han, Kai Chen, Chao Gong, Zhipeng Wei, Jingjing Chen, Yu-Gang Jiang

In contrast to previous methods, DuMo employs the Eraser with PRior Knowledge (EPR) module which modifies the skip connection features of the U-NET and primarily achieves concept erasure on details (high-frequency) components of the image.

AIM: Additional Image Guided Generation of Transferable Adversarial Attacks

no code implementations2 Jan 2025 Teng Li, Xingjun Ma, Yu-Gang Jiang

In this work, we focus on generative approaches for targeted transferable attacks.

4D Gaussian Splatting: Modeling Dynamic Scenes with Native 4D Primitives

no code implementations30 Dec 2024 Zeyu Yang, Zijie Pan, Xiatian Zhu, Li Zhang, Yu-Gang Jiang, Philip H. S. Torr

Dynamic 3D scene representation and novel view synthesis from captured videos are crucial for enabling immersive experiences required by AR/VR and metaverse applications.

Novel View Synthesis Scene Understanding

STNMamba: Mamba-based Spatial-Temporal Normality Learning for Video Anomaly Detection

no code implementations28 Dec 2024 Zhangxun Li, Mengyang Zhao, Xuan Yang, Yang Liu, Jiamu Sheng, Xinhua Zeng, Tian Wang, Kewei Wu, Yu-Gang Jiang

Within this module, the Spatial-Temporal Fusion Block (STFB) is proposed to fuse the spatial and temporal features into a unified feature space, and the memory bank is utilized to store spatial-temporal prototypes of normal patterns, restricting the model's ability to represent anomalies.

Anomaly Detection Mamba +1

CreatiLayout: Siamese Multimodal Diffusion Transformer for Creative Layout-to-Image Generation

no code implementations5 Dec 2024 HUI ZHANG, Dexiang Hong, Tingwei Gao, Yitong Wang, Jie Shao, Xinglong Wu, Zuxuan Wu, Yu-Gang Jiang

To Inherit the advantages of MM-DiT, we use a separate set of network weights to process the layout, treating it as equally important as the image and text modalities.

Layout Generation Layout-to-Image Generation

Inst-IT: Boosting Multimodal Instance Understanding via Explicit Visual Prompt Instruction Tuning

1 code implementation4 Dec 2024 Wujian Peng, Lingchen Meng, Yitong Chen, Yiweng Xie, Yang Liu, Tao Gui, Hang Xu, Xipeng Qiu, Zuxuan Wu, Yu-Gang Jiang

Building upon this pipeline, we proposed Inst-IT, a solution to enhance LMMs in Instance understanding via explicit visual prompt Instruction Tuning.

Ranked #3 on Visual Question Answering on ViP-Bench (using extra training data)

Multimodal Large Language Model Video Understanding +3

SparseGrasp: Robotic Grasping via 3D Semantic Gaussian Splatting from Sparse Multi-View RGB Images

no code implementations3 Dec 2024 Junqiu Yu, Xinlin Ren, Yongchong Gu, Haitao Lin, Tianyu Wang, Yi Zhu, Hang Xu, Yu-Gang Jiang, xiangyang xue, Yanwei Fu

Language-guided robotic grasping is a rapidly advancing field where robots are instructed using human language to grasp specific objects.

3DGS Robotic Grasping

DiffPatch: Generating Customizable Adversarial Patches using Diffusion Model

2 code implementations2 Dec 2024 Zhixiang Wang, Guangnan Ye, Xiaosen Wang, Siheng Chen, Zhibo Wang, Xingjun Ma, Yu-Gang Jiang

However, most existing adversarial patch generation methods prioritize attack effectiveness over stealthiness, resulting in patches that are aesthetically unpleasing.

model

ForgerySleuth: Empowering Multimodal Large Language Models for Image Manipulation Detection

1 code implementation29 Nov 2024 Zhihao Sun, Haoran Jiang, Haoran Chen, Yixin Cao, Xipeng Qiu, Zuxuan Wu, Yu-Gang Jiang

Moreover, we construct the ForgeryAnalysis dataset through the Chain-of-Clues prompt, which includes analysis and reasoning text to upgrade the image manipulation detection task.

Image Manipulation Image Manipulation Detection

LoRA of Change: Learning to Generate LoRA for the Editing Instruction from A Single Before-After Image Pair

no code implementations28 Nov 2024 Xue Song, Jiequan Cui, Hanwang Zhang, Jiaxin Shi, Jingjing Chen, Chi Zhang, Yu-Gang Jiang

Furthermore, generalizable models for image editing with visual instructions typically require quad data, i. e., a before-after image pair, along with query and target images.

Specificity Text-based Image Editing

SVTRv2: CTC Beats Encoder-Decoder Models in Scene Text Recognition

1 code implementation24 Nov 2024 Yongkun Du, Zhineng Chen, Hongtao Xie, Caiyan Jia, Yu-Gang Jiang

In this paper, we propose SVTRv2, a CTC model that beats leading EDTRs in both accuracy and inference speed.

Decoder Optical Character Recognition (OCR) +1

REDUCIO! Generating 1024$\times$1024 Video within 16 Seconds using Extremely Compressed Motion Latents

1 code implementation20 Nov 2024 Rui Tian, Qi Dai, Jianmin Bao, Kai Qiu, Yifan Yang, Chong Luo, Zuxuan Wu, Yu-Gang Jiang

Commercial video generation models have exhibited realistic, high-fidelity results but are still restricted to limited access.

GPU Video Generation

Visual Cue Enhancement and Dual Low-Rank Adaptation for Efficient Visual Instruction Fine-Tuning

no code implementations19 Nov 2024 Pengkun Jiao, Bin Zhu, Jingjing Chen, Chong-Wah Ngo, Yu-Gang Jiang

Parameter-efficient fine-tuning multimodal large language models (MLLMs) presents significant challenges, including reliance on high-level visual features that limit fine-grained detail comprehension, and data conflicts that arise from task complexity.

parameter-efficient fine-tuning

Retrieval Augmented Recipe Generation

no code implementations13 Nov 2024 Guoshan Liu, Hailong Yin, Bin Zhu, Jingjing Chen, Chong-Wah Ngo, Yu-Gang Jiang

Existing works for recipe generation primarily utilize a two-stage training method, first generating ingredients and then obtaining instructions from both the image and ingredients.

Recipe Generation Retrieval

Domain Expansion and Boundary Growth for Open-Set Single-Source Domain Generalization

no code implementations5 Nov 2024 Pengkun Jiao, Na Zhao, Jingjing Chen, Yu-Gang Jiang

In this paper, we propose a novel learning approach based on domain expansion and boundary growth to expand the scarce source samples and enlarge the boundaries across the known classes that indirectly broaden the boundary between the known and unknown classes.

image-classification Image Classification +1

IDEATOR: Jailbreaking Large Vision-Language Models Using Themselves

no code implementations29 Oct 2024 Ruofan Wang, Bo wang, Xiaosen Wang, Xingjun Ma, Yu-Gang Jiang

Specifically, IDEATOR uses a VLM to create targeted jailbreak texts and pairs them with jailbreak images generated by a state-of-the-art diffusion model.

Expose Before You Defend: Unifying and Enhancing Backdoor Defenses via Exposed Models

1 code implementation25 Oct 2024 Yige Li, Hanxun Huang, Jiaming Zhang, Xingjun Ma, Yu-Gang Jiang

Specifically, EBYD first exposes the backdoor functionality in the backdoored model through a model preprocessing step called backdoor exposure, and then applies detection and removal methods to the exposed model to identify and eliminate the backdoor features.

backdoor defense Model Editing +1

UnSeg: One Universal Unlearnable Example Generator is Enough against All Image Segmentation

1 code implementation13 Oct 2024 Ye Sun, Hao Zhang, Tiehua Zhang, Xingjun Ma, Yu-Gang Jiang

In this work, we exploit the concept of unlearnable examples to make images unusable to model training by generating and adding unlearnable noise into the original images.

All Bilevel Optimization +4

EAGLE: Towards Efficient Arbitrary Referring Visual Prompts Comprehension for Multimodal Large Language Models

no code implementations25 Sep 2024 Jiacheng Zhang, Yang Jiao, Shaoxiang Chen, Jingjing Chen, Yu-Gang Jiang

To effectively instruct an MLLM, in addition to conventional language expressions, the practice of referring to objects by painting with brushes on images has emerged as a prevalent tool (referred to as "referring visual prompts") due to its efficacy in aligning the user's intention with specific image regions.

Instruction Following

FreeEnhance: Tuning-Free Image Enhancement via Content-Consistent Noising-and-Denoising Process

no code implementations11 Sep 2024 Yang Luo, Yiheng Zhang, Zhaofan Qiu, Ting Yao, Zhineng Chen, Yu-Gang Jiang, Tao Mei

Technically, FreeEnhance is a two-stage process that firstly adds random noise to the input image and then capitalizes on a pre-trained image diffusion model (i. e., Latent Diffusion Models) to denoise and enhance the image details.

Denoising Image Enhancement +2

DreamMesh: Jointly Manipulating and Texturing Triangle Meshes for Text-to-3D Generation

no code implementations11 Sep 2024 Haibo Yang, Yang Chen, Yingwei Pan, Ting Yao, Zhineng Chen, Zuxuan Wu, Yu-Gang Jiang, Tao Mei

In the fine stage, DreamMesh jointly manipulates the mesh and refines the texture map, leading to high-quality triangle meshes with high-fidelity textured materials.

3D Architecture 3D Generation +2

GenRec: Unifying Video Generation and Recognition with Diffusion Models

1 code implementation27 Aug 2024 Zejia Weng, Xitong Yang, Zhen Xing, Zuxuan Wu, Yu-Gang Jiang

In this paper, we aim to investigate whether such priors derived from a generative process are suitable for video recognition, and eventually joint optimization of generation and recognition.

Image to Video Generation Video Recognition

Decoder Pre-Training with only Text for Scene Text Recognition

1 code implementation11 Aug 2024 Shuai Zhao, Yongkun Du, Zhineng Chen, Yu-Gang Jiang

Extensive experiments across various STR decoders and language recognition tasks underscore the broad applicability and remarkable performance of DPTR, providing a novel insight for STR pre-training.

Decoder Scene Text Recognition

ReToMe-VA: Recursive Token Merging for Video Diffusion-based Unrestricted Adversarial Attack

no code implementations10 Aug 2024 Ziyi Gao, Kai Chen, Zhipeng Wei, Tingshu Mou, Jingjing Chen, Zhiyu Tan, Hao Li, Yu-Gang Jiang

However, existing works on diffusion-based unrestricted attacks are mostly focused on images yet are seldom explored in videos.

Adversarial Attack Denoising

EnJa: Ensemble Jailbreak on Large Language Models

no code implementations7 Aug 2024 Jiahao Zhang, Zilong Wang, Ruofan Wang, Xingjun Ma, Yu-Gang Jiang

As Large Language Models (LLMs) are increasingly being deployed in safety-critical applications, their vulnerability to potential jailbreaks -- malicious prompts that can disable the safety mechanism of LLMs -- has attracted growing research attention.

Safety Alignment

AdvQDet: Detecting Query-Based Adversarial Attacks with Adversarial Contrastive Prompt Tuning

1 code implementation4 Aug 2024 Xin Wang, Kai Chen, Xingjun Ma, Zhineng Chen, Jingjing Chen, Yu-Gang Jiang

During this process, the queries made to the target model are intermediate adversarial examples crafted at the previous attack step, which share high similarities in the pixel space.

Downstream Transfer Attack: Adversarial Attacks on Downstream Models with Pre-trained Vision Transformers

no code implementations3 Aug 2024 Weijie Zheng, Xingjun Ma, Hanxun Huang, Zuxuan Wu, Yu-Gang Jiang

With the advancement of vision transformers (ViTs) and self-supervised learning (SSL) techniques, pre-trained large ViTs have become the new foundation models for computer vision applications.

Self-Supervised Learning

Reliable and Efficient Concept Erasure of Text-to-Image Diffusion Models

1 code implementation17 Jul 2024 Chao Gong, Kai Chen, Zhipeng Wei, Jingjing Chen, Yu-Gang Jiang

In this work, we introduce Reliable and Efficient Concept Erasure (RECE), a novel approach that modifies the model in 3 seconds without necessitating additional fine-tuning.

Benchmarking Red Teaming

Infinite Motion: Extended Motion Generation via Long Text Instructions

no code implementations11 Jul 2024 Mengtian Li, Chengshuo Zhai, Shengxiang Yao, Zhifeng Xie, Keyu Chen, Yu-Gang Jiang

We further demonstrate the versatility and practical utility of "Infinite Motion" through three specific applications: natural language interactive editing, motion sequence editing within long sequences and splicing of independent motion sequences.

Motion Generation Motion Synthesis

Unlocking Textual and Visual Wisdom: Open-Vocabulary 3D Object Detection Enhanced by Comprehensive Guidance from Text and Image

no code implementations7 Jul 2024 Pengkun Jiao, Na Zhao, Jingjing Chen, Yu-Gang Jiang

Open-vocabulary 3D object detection (OV-3DDet) aims to localize and recognize both seen and previously unseen object categories within any new 3D scene.

3D Object Detection Object +1

PECTP: Parameter-Efficient Cross-Task Prompts for Incremental Vision Transformer

1 code implementation4 Jul 2024 Qian Feng, Hanbin Zhao, Chao Zhang, Jiahua Dong, Henghui Ding, Yu-Gang Jiang, Hui Qian

Prompt-fixed methods only learn a single set of prompts on one of the incremental tasks and can not handle all the incremental tasks effectively.

Incremental Learning

A Survey of Multimodal-Guided Image Editing with Text-to-Image Diffusion Models

1 code implementation20 Jun 2024 Xincheng Shuai, Henghui Ding, Xingjun Ma, RongCheng Tu, Yu-Gang Jiang, DaCheng Tao

Image editing aims to edit the given synthetic or real image to meet the specific requirements from users.

Video Editing

Extracting Training Data from Unconditional Diffusion Models

no code implementations18 Jun 2024 Yunhao Chen, Xingjun Ma, Difan Zou, Yu-Gang Jiang

In this work, we aim to establish a theoretical understanding of memorization in DPMs with 1) a memorization metric for theoretical analysis, 2) an analysis of conditional memorization with informative and random labels, and 3) two better evaluation metrics for measuring memorization.

Memorization

OmniTokenizer: A Joint Image-Video Tokenizer for Visual Generation

1 code implementation13 Jun 2024 Junke Wang, Yi Jiang, Zehuan Yuan, Binyue Peng, Zuxuan Wu, Yu-Gang Jiang

To exploit the complementary nature of image and video data, we further propose a progressive training strategy, where OmniTokenizer is first trained on image data on a fixed resolution to develop the spatial encoding capacity and then jointly trained on image and video data on multiple resolutions to learn the temporal dynamics.

Video Generation Video Prediction

AID: Adapting Image2Video Diffusion Models for Instruction-guided Video Prediction

no code implementations10 Jun 2024 Zhen Xing, Qi Dai, Zejia Weng, Zuxuan Wu, Yu-Gang Jiang

Text-guided video prediction (TVP) involves predicting the motion of future frames from the initial frame according to an instruction, which has wide applications in virtual reality, robotics, and content creation.

Language Modelling Large Language Model +1

DeepStack: Deeply Stacking Visual Tokens is Surprisingly Simple and Effective for LMMs

no code implementations6 Jun 2024 Lingchen Meng, Jianwei Yang, Rui Tian, Xiyang Dai, Zuxuan Wu, Jianfeng Gao, Yu-Gang Jiang

The resulting architecture is simple but significantly increases computation and memory costs, as it has to handle a large number of additional tokens in its input layer.

Language Modelling Large Language Model +3

MotionFollower: Editing Video Motion via Lightweight Score-Guided Diffusion

1 code implementation30 May 2024 Shuyuan Tu, Qi Dai, Zihao Zhang, Sicheng Xie, Zhi-Qi Cheng, Chong Luo, Xintong Han, Zuxuan Wu, Yu-Gang Jiang

In this paper, we propose MotionFollower, a lightweight score-guided diffusion model for video motion editing.

Denoising GPU +3

White-box Multimodal Jailbreaks Against Large Vision-Language Models

1 code implementation28 May 2024 Ruofan Wang, Xingjun Ma, Hanxu Zhou, Chuanjun Ji, Guangnan Ye, Yu-Gang Jiang

Subsequently, an adversarial text suffix is integrated and co-optimized with the adversarial image prefix to maximize the probability of eliciting affirmative responses to various harmful instructions.

Adversarial Robustness Adversarial Text

ModelLock: Locking Your Model With a Spell

no code implementations25 May 2024 Yifeng Gao, Yuhua Sun, Xingjun Ma, Zuxuan Wu, Yu-Gang Jiang

This paper presents a novel model protection paradigm ModelLock that locks (destroys) the performance of a model on normal clean data so as to make it unusable or unextractable without the right key.

image-classification Image Classification +2

Adaptive Retention & Correction: Test-Time Training for Continual Learning

no code implementations23 May 2024 Haoran Chen, Micah Goldblum, Zuxuan Wu, Yu-Gang Jiang

A common problem in continual learning is the classification layer's bias towards the most recent task.

ARC Continual Learning +2

FedCAda: Adaptive Client-Side Optimization for Accelerated and Stable Federated Learning

no code implementations20 May 2024 Liuzhi Zhou, Yu He, Kun Zhai, Xiang Liu, Sen Liu, Xingjun Ma, Guangnan Ye, Yu-Gang Jiang, Hongfeng Chai

This comparative analysis revealed that due to the limited information contained within client models from other clients during the initial stages of federated learning, more substantial constraints need to be imposed on the parameters of the adaptive algorithm.

Federated Learning

Zero-shot High-fidelity and Pose-controllable Character Animation

no code implementations21 Apr 2024 Bingwen Zhu, Fanyi Wang, Tianyi Lu, Peng Liu, Jingwen Su, Jinxiu Liu, Yanhao Zhang, Zuxuan Wu, Guo-Jun Qi, Yu-Gang Jiang

Image-to-video (I2V) generation aims to create a video sequence from a single image, which requires high temporal coherence and visual fidelity.

Look Before You Decide: Prompting Active Deduction of MLLMs for Assumptive Reasoning

no code implementations19 Apr 2024 Yian Li, Wentao Tian, Yang Jiao, Jingjing Chen, Tianwen Qian, Bin Zhu, Na Zhao, Yu-Gang Jiang

Recently, Multimodal Large Language Models (MLLMs) have achieved significant success across multiple disciplines due to their exceptional instruction-following capabilities and extensive world knowledge.

Benchmarking counterfactual +5

FedEGG: Federated Learning with Explicit Global Guidance

no code implementations18 Apr 2024 Kun Zhai, Yifeng Gao, Difan Zou, Guangnan Ye, Siheng Chen, Xingjun Ma, Yu-Gang Jiang

Federated Learning (FL) holds great potential for diverse applications owing to its privacy-preserving nature.

Federated Learning Privacy Preserving

Learning to Rank Patches for Unbiased Image Redundancy Reduction

1 code implementation CVPR 2024 Yang Luo, Zhineng Chen, Peng Zhou, Zuxuan Wu, Xieping Gao, Yu-Gang Jiang

The results demonstrate that LTRP outperforms both supervised and other self-supervised methods due to the fair assessment of image content.

Image Reconstruction Inductive Bias +1

OmniVid: A Generative Framework for Universal Video Understanding

1 code implementation CVPR 2024 Junke Wang, Dongdong Chen, Chong Luo, Bo He, Lu Yuan, Zuxuan Wu, Yu-Gang Jiang

The core of video understanding tasks, such as recognition, captioning, and tracking, is to automatically detect objects or actions in a video and analyze their temporal evolution.

Action Recognition Decoder +5

Whose Side Are You On? Investigating the Political Stance of Large Language Models

1 code implementation15 Mar 2024 Pagnarasmey Pit, Xingjun Ma, Mike Conway, Qingyu Chen, James Bailey, Henry Pit, Putrasmey Keo, Watey Diep, Yu-Gang Jiang

Large Language Models (LLMs) have gained significant popularity for their application in various everyday tasks such as text generation, summarization, and information retrieval.

Fairness Information Retrieval +1

From Canteen Food to Daily Meals: Generalizing Food Recognition to More Practical Scenarios

no code implementations12 Mar 2024 Guoshan Liu, Yang Jiao, Jingjing Chen, Bin Zhu, Yu-Gang Jiang

These two datasets are used to evaluate the transferability of approaches from the well-curated food image domain to the everyday-life food image domain.

Food Recognition

Lumen: Unleashing Versatile Vision-Centric Capabilities of Large Multimodal Models

1 code implementation12 Mar 2024 Yang Jiao, Shaoxiang Chen, Zequn Jie, Jingjing Chen, Lin Ma, Yu-Gang Jiang

To address this issue, we propose a novel LMM architecture named Lumen, a Large multimodal model with versatile vision-centric capability enhancement.

Concept Alignment Instruction Following +2

Doubly Abductive Counterfactual Inference for Text-based Image Editing

1 code implementation CVPR 2024 Xue Song, Jiequan Cui, Hanwang Zhang, Jingjing Chen, Richang Hong, Yu-Gang Jiang

Through the lens of the formulation, we find that the crux of TBIE is that existing techniques hardly achieve a good trade-off between editability and fidelity, mainly due to the overfitting of the single-image fine-tuning.

counterfactual Counterfactual Inference +2

Instruction-Guided Scene Text Recognition

1 code implementation31 Jan 2024 Yongkun Du, Zhineng Chen, Yuchen Su, Caiyan Jia, Yu-Gang Jiang

We propose a novel instruction-guided scene text recognition (IGTR) paradigm that formulates STR as an instruction learning problem and understands text images by predicting character attributes, e. g., character frequency, position, etc.

Question Answering Scene Text Recognition

Shortcuts Everywhere and Nowhere: Exploring Multi-Trigger Backdoor Attacks

1 code implementation27 Jan 2024 Yige Li, Jiabo He, Hanxun Huang, Jun Sun, Xingjun Ma, Yu-Gang Jiang

Backdoor attacks have become a significant threat to the pre-training and deployment of deep neural networks (DNNs).

FoodLMM: A Versatile Food Assistant using Large Multi-modal Model

no code implementations22 Dec 2023 Yuehao Yin, Huiyan Qi, Bin Zhu, Jingjing Chen, Yu-Gang Jiang, Chong-Wah Ngo

In the second stage, we construct a multi-round conversation dataset and a reasoning segmentation dataset to fine-tune the model, enabling it to conduct professional dialogues and generate segmentation masks based on complex reasoning in the food domain.

Food Recognition Multi-Task Learning +4

MotionEditor: Editing Video Motion via Content-Aware Diffusion

1 code implementation CVPR 2024 Shuyuan Tu, Qi Dai, Zhi-Qi Cheng, Han Hu, Xintong Han, Zuxuan Wu, Yu-Gang Jiang

This mechanism enables the editing branch to query the key and value from the reconstruction branch in a decoupled manner, making the editing branch retain the original background and protagonist appearance.

Video Editing

MagDiff: Multi-Alignment Diffusion for High-Fidelity Video Generation and Editing

1 code implementation29 Nov 2023 Haoyu Zhao, Tianyi Lu, Jiaxi Gu, Xing Zhang, Qingping Zheng, Zuxuan Wu, Hang Xu, Yu-Gang Jiang

The high-fidelity alignment is developed to further enhance the fidelity of both video generation and editing by taking the subject image as an additional model input.

Denoising Image to Video Generation +1

AdaDiff: Adaptive Step Selection for Fast Diffusion Models

no code implementations24 Nov 2023 HUI ZHANG, Zuxuan Wu, Zhen Xing, Jie Shao, Yu-Gang Jiang

Diffusion models, as a type of generative model, have achieved impressive results in generating images and videos conditioned on textual conditions.

Denoising Image Generation +1

SEGIC: Unleashing the Emergent Correspondence for In-Context Segmentation

1 code implementation24 Nov 2023 Lingchen Meng, Shiyi Lan, Hengduo Li, Jose M. Alvarez, Zuxuan Wu, Yu-Gang Jiang

In-context segmentation aims at segmenting novel images using a few labeled example images, termed as "in-context examples", exploring content similarities between examples and the target.

Meta-Learning One-Shot Segmentation +3

Adversarial Prompt Tuning for Vision-Language Models

1 code implementation19 Nov 2023 Jiaming Zhang, Xingjun Ma, Xin Wang, Lingyu Qiu, Jiaqi Wang, Yu-Gang Jiang, Jitao Sang

With the rapid advancement of multimodal learning, pre-trained Vision-Language Models (VLMs) such as CLIP have demonstrated remarkable capacities in bridging the gap between visual and language modalities.

Adversarial Robustness

To See is to Believe: Prompting GPT-4V for Better Visual Instruction Tuning

2 code implementations13 Nov 2023 Junke Wang, Lingchen Meng, Zejia Weng, Bo He, Zuxuan Wu, Yu-Gang Jiang

Existing visual instruction tuning methods typically prompt large language models with textual descriptions to generate instruction-following data.

Instruction Following MM-Vet +1

Fake Alignment: Are LLMs Really Aligned Well?

1 code implementation10 Nov 2023 Yixu Wang, Yan Teng, Kexin Huang, Chengqi Lyu, Songyang Zhang, Wenwei Zhang, Xingjun Ma, Yu-Gang Jiang, Yu Qiao, Yingchun Wang

The growing awareness of safety concerns in large language models (LLMs) has sparked considerable interest in the evaluation of safety.

Multiple-choice

A Survey on Video Diffusion Models

1 code implementation16 Oct 2023 Zhen Xing, Qijun Feng, Haoran Chen, Qi Dai, Han Hu, Hang Xu, Zuxuan Wu, Yu-Gang Jiang

However, existing surveys mainly focus on diffusion models in the context of image generation, with few up-to-date reviews on their application in the video domain.

Image Generation Survey +3

Building an Open-Vocabulary Video CLIP Model with Better Architectures, Optimization and Data

1 code implementation8 Oct 2023 Zuxuan Wu, Zejia Weng, Wujian Peng, Xitong Yang, Ang Li, Larry S. Davis, Yu-Gang Jiang

Despite significant results achieved by Contrastive Language-Image Pretraining (CLIP) in zero-shot image recognition, limited effort has been made exploring its potential for zero-shot video recognition.

Action Recognition Continual Learning +4

Reuse and Diffuse: Iterative Denoising for Text-to-Video Generation

1 code implementation7 Sep 2023 Jiaxi Gu, Shicong Wang, Haoyu Zhao, Tianyi Lu, Xing Zhang, Zuxuan Wu, Songcen Xu, Wei zhang, Yu-Gang Jiang, Hang Xu

Conditioned on an initial video clip with a small number of frames, additional frames are iteratively generated by reusing the original latent features and following the previous diffusion process.

Action Recognition Decoder +4

SimDA: Simple Diffusion Adapter for Efficient Video Generation

no code implementations CVPR 2024 Zhen Xing, Qi Dai, Han Hu, Zuxuan Wu, Yu-Gang Jiang

In this work, we propose a Simple Diffusion Adapter (SimDA) that fine-tunes only 24M out of 1. 1B parameters of a strong T2I model, adapting it to video generation in a parameter-efficient way.

Transfer Learning Video Editing +2

On the Importance of Spatial Relations for Few-shot Action Recognition

no code implementations14 Aug 2023 Yilun Zhang, Yuqian Fu, Xingjun Ma, Lizhe Qi, Jingjing Chen, Zuxuan Wu, Yu-Gang Jiang

We are thus motivated to investigate the importance of spatial relations and propose a more accurate few-shot action recognition method that leverages both spatial and temporal information.

Few-Shot action recognition Few Shot Action Recognition +1

Context Perception Parallel Decoder for Scene Text Recognition

2 code implementations23 Jul 2023 Yongkun Du, Zhineng Chen, Caiyan Jia, Xiaoting Yin, Chenxia Li, Yuning Du, Yu-Gang Jiang

We first present an empirical study of AR decoding in STR, and discover that the AR decoder not only models linguistic context, but also provides guidance on visual context perception.

 Ranked #1 on Scene Text Recognition on CUTE80 (using extra training data)

Decoder Language Modelling +1

Prompting Large Language Models to Reformulate Queries for Moment Localization

no code implementations6 Jun 2023 Wenfeng Yan, Shaoxiang Chen, Zuxuan Wu, Yu-Gang Jiang

The task of moment localization is to localize a temporal moment in an untrimmed video for a given natural language query.

Moment Queries Natural Language Queries

Reconstructive Neuron Pruning for Backdoor Defense

1 code implementation24 May 2023 Yige Li, Xixiang Lyu, Xingjun Ma, Nodens Koren, Lingjuan Lyu, Bo Li, Yu-Gang Jiang

Specifically, RNP first unlearns the neurons by maximizing the model's error on a small subset of clean samples and then recovers the neurons by minimizing the model's error on the same data.

backdoor defense

MRN: Multiplexed Routing Network for Incremental Multilingual Text Recognition

1 code implementation ICCV 2023 Tianlun Zheng, Zhineng Chen, Bingchen Huang, Wei zhang, Yu-Gang Jiang

In this paper, we propose the Incremental MLTR (IMLTR) task in the context of incremental learning (IL), where different languages are introduced in batches.

Continual Learning Incremental Learning +2

NuScenes-QA: A Multi-modal Visual Question Answering Benchmark for Autonomous Driving Scenario

1 code implementation24 May 2023 Tianwen Qian, Jingjing Chen, Linhai Zhuo, Yang Jiao, Yu-Gang Jiang

We introduce a novel visual question answering (VQA) task in the context of autonomous driving, aiming to answer natural language questions based on street-view clues.

Autonomous Driving Question Answering +1

TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition

1 code implementation9 May 2023 Tianlun Zheng, Zhineng Chen, Jinfeng Bai, Hongtao Xie, Yu-Gang Jiang

In this work, we introduce TPS++, an attention-enhanced TPS transformation that incorporates the attention mechanism to text rectification for the first time.

Optical Character Recognition (OCR) Scene Text Recognition

ChatVideo: A Tracklet-centric Multimodal and Versatile Video Understanding System

no code implementations27 Apr 2023 Junke Wang, Dongdong Chen, Chong Luo, Xiyang Dai, Lu Yuan, Zuxuan Wu, Yu-Gang Jiang

Existing deep video models are limited by specific tasks, fixed input-output spaces, and poor generalization capabilities, making it difficult to deploy them in real-world scenarios.

Video Understanding

Implicit Temporal Modeling with Learnable Alignment for Video Recognition

1 code implementation ICCV 2023 Shuyuan Tu, Qi Dai, Zuxuan Wu, Zhi-Qi Cheng, Han Hu, Yu-Gang Jiang

While modeling temporal information within straight through tube is widely adopted in literature, we find that simple frame alignment already provides enough essence without temporal attention.

Action Classification Action Recognition +1

DiffusionAD: Norm-guided One-step Denoising Diffusion for Anomaly Detection

2 code implementations15 Mar 2023 HUI ZHANG, Zheng Wang, Dan Zeng, Zuxuan Wu, Yu-Gang Jiang

We introduce DiffusionAD, a novel anomaly detection pipeline comprising a reconstruction sub-network and a segmentation sub-network.

Denoising Unsupervised Anomaly Detection

PromptFusion: Decoupling Stability and Plasticity for Continual Learning

1 code implementation13 Mar 2023 Haoran Chen, Zuxuan Wu, Xintong Han, Menglin Jia, Yu-Gang Jiang

Current research on continual learning mainly focuses on relieving catastrophic forgetting, and most of their success is at the cost of limiting the performance of newly incoming tasks.

class-incremental learning Class Incremental Learning +1

StyleAdv: Meta Style Adversarial Training for Cross-Domain Few-Shot Learning

2 code implementations CVPR 2023 Yuqian Fu, Yu Xie, Yanwei Fu, Yu-Gang Jiang

Thus, inspired by vanilla adversarial learning, a novel model-agnostic meta Style Adversarial training (StyleAdv) method together with a novel style adversarial attack method is proposed for CD-FSL.

Adversarial Attack cross-domain few-shot learning

Open-VCLIP: Transforming CLIP to an Open-vocabulary Video Model via Interpolated Weight Optimization

1 code implementation1 Feb 2023 Zejia Weng, Xitong Yang, Ang Li, Zuxuan Wu, Yu-Gang Jiang

Our framework extends CLIP with minimal modifications to model spatial-temporal relationships in videos, making it a specialized video classifier, while striving for generalization.

Action Recognition Continual Learning +2

Vocabulary-informed Zero-shot and Open-set Learning

1 code implementation3 Jan 2023 Yanwei Fu, Xiaomei Wang, Hanze Dong, Yu-Gang Jiang, Meng Wang, xiangyang xue, Leonid Sigal

Despite significant progress in object categorization, in recent years, a number of important challenges remain; mainly, the ability to learn from limited labeled data and to recognize object classes within large, potentially open, set of labels.

Object Categorization Open Set Learning +1

Bi-Directional Feature Fusion Generative Adversarial Network for Ultra-High Resolution Pathological Image Virtual Re-Staining

no code implementations CVPR 2023 Kexin Sun, Zhineng Chen, Gongwei Wang, Jun Liu, Xiongjun Ye, Yu-Gang Jiang

In order to eliminate the square effect, we design a bi-directional feature fusion generative adversarial network (BFF-GAN) with a global branch and a local branch.

Generative Adversarial Network

Unlearnable Clusters: Towards Label-agnostic Unlearnable Examples

1 code implementation CVPR 2023 Jiaming Zhang, Xingjun Ma, Qi Yi, Jitao Sang, Yu-Gang Jiang, YaoWei Wang, Changsheng Xu

Furthermore, we propose to leverage VisionandLanguage Pre-trained Models (VLPMs) like CLIP as the surrogate model to improve the transferability of the crafted UCs to diverse domains.

Data Poisoning

Look Before You Match: Instance Understanding Matters in Video Object Segmentation

no code implementations CVPR 2023 Junke Wang, Dongdong Chen, Zuxuan Wu, Chong Luo, Chuanxin Tang, Xiyang Dai, Yucheng Zhao, Yujia Xie, Lu Yuan, Yu-Gang Jiang

Towards this goal, we present a two-branch network for VOS, where the query-based instance segmentation (IS) branch delves into the instance details of the current frame and the VOS branch performs spatial-temporal matching with the memory bank.

Instance Segmentation Segmentation +3

Masked Video Distillation: Rethinking Masked Feature Modeling for Self-supervised Video Representation Learning

4 code implementations CVPR 2023 Rui Wang, Dongdong Chen, Zuxuan Wu, Yinpeng Chen, Xiyang Dai, Mengchen Liu, Lu Yuan, Yu-Gang Jiang

For the choice of teacher models, we observe that students taught by video teachers perform better on temporally-heavy video tasks, while image teachers transfer stronger spatial representations for spatially-heavy video tasks.

Action Classification Representation Learning +1

Prototypical Residual Networks for Anomaly Detection and Localization

no code implementations CVPR 2023 HUI ZHANG, Zuxuan Wu, Zheng Wang, Zhineng Chen, Yu-Gang Jiang

Anomaly detection and localization are widely used in industrial manufacturing for its efficiency and effectiveness.

Ranked #4 on Supervised Anomaly Detection on MVTec AD (using extra training data)

Supervised Anomaly Detection

ResFormer: Scaling ViTs with Multi-Resolution Training

1 code implementation CVPR 2023 Rui Tian, Zuxuan Wu, Qi Dai, Han Hu, Yu Qiao, Yu-Gang Jiang

We introduce, ResFormer, a framework that is built upon the seminal idea of multi-resolution training for improved performance on a wide spectrum of, mostly unseen, testing resolutions.

Action Recognition image-classification +5

Transferability Estimation Based On Principal Gradient Expectation

no code implementations29 Nov 2022 Huiyan Qi, Lechao Cheng, Jingjing Chen, Yue Yu, Xue Song, Zunlei Feng, Yu-Gang Jiang

Transfer learning aims to improve the performance of target tasks by transferring knowledge acquired in source tasks.

Transfer Learning

TGDM: Target Guided Dynamic Mixup for Cross-Domain Few-Shot Learning

1 code implementation11 Oct 2022 Linhai Zhuo, Yuqian Fu, Jingjing Chen, Yixin Cao, Yu-Gang Jiang

The proposed TGDM framework contains a Mixup-3T network for learning classifiers and a dynamic ratio generation network (DRGN) for learning the optimal mix ratio.

cross-domain few-shot learning Transfer Learning

ME-D2N: Multi-Expert Domain Decompositional Network for Cross-Domain Few-Shot Learning

1 code implementation11 Oct 2022 Yuqian Fu, Yu Xie, Yanwei Fu, Jingjing Chen, Yu-Gang Jiang

Concretely, to solve the data imbalance problem between the source data with sufficient examples and the auxiliary target data with limited examples, we build our model under the umbrella of multi-expert learning.

cross-domain few-shot learning Knowledge Distillation

Text-driven Video Prediction

no code implementations6 Oct 2022 Xue Song, Jingjing Chen, Bin Zhu, Yu-Gang Jiang

Specifically, appearance and motion components are provided by the image and caption separately.

Causal Inference Prediction +2

Locate before Answering: Answer Guided Question Localization for Video Question Answering

no code implementations5 Oct 2022 Tianwen Qian, Ran Cui, Jingjing Chen, Pai Peng, Xiaowei Guo, Yu-Gang Jiang

Considering the fact that the question often remains concentrated in a short temporal range, we propose to first locate the question to a segment in the video and then infer the answer using the located segment only.

Question Answering Video Question Answering

Semi-Supervised Single-View 3D Reconstruction via Prototype Shape Priors

1 code implementation30 Sep 2022 Zhen Xing, Hengduo Li, Zuxuan Wu, Yu-Gang Jiang

In particular, we introduce an attention-guided prototype shape prior module for guiding realistic object reconstruction.

3D Reconstruction image-classification +3

OmniVL:One Foundation Model for Image-Language and Video-Language Tasks

no code implementations15 Sep 2022 Junke Wang, Dongdong Chen, Zuxuan Wu, Chong Luo, Luowei Zhou, Yucheng Zhao, Yujia Xie, Ce Liu, Yu-Gang Jiang, Lu Yuan

This paper presents OmniVL, a new foundation model to support both image-language and video-language tasks using one universal architecture.

Ranked #4 on Cross-Modal Retrieval on Flickr30k (using extra training data)

Action Classification Action Recognition +15

Enhancing the Self-Universality for Transferable Targeted Attacks

1 code implementation CVPR 2023 Zhipeng Wei, Jingjing Chen, Zuxuan Wu, Yu-Gang Jiang

Our new attack method is proposed based on the observation that highly universal adversarial perturbations tend to be more transferable for targeted attacks.

MSMDFusion: Fusing LiDAR and Camera at Multiple Scales with Multi-Depth Seeds for 3D Object Detection

1 code implementation CVPR 2023 Yang Jiao, Zequn Jie, Shaoxiang Chen, Jingjing Chen, Lin Ma, Yu-Gang Jiang

Recent approaches aim at exploring the semantic densities of camera features through lifting points in 2D camera images (referred to as seeds) into 3D space, and then incorporate 2D semantics via cross-modal interaction or fusion techniques.

3D Object Detection Autonomous Driving +1

Video Mobile-Former: Video Recognition with Efficient Global Spatial-temporal Modeling

no code implementations25 Aug 2022 Rui Wang, Zuxuan Wu, Dongdong Chen, Yinpeng Chen, Xiyang Dai, Mengchen Liu, Luowei Zhou, Lu Yuan, Yu-Gang Jiang

To avoid significant computational cost incurred by computing self-attention between the large number of local patches in videos, we propose to use very few global tokens (e. g., 6) for a whole video in Transformers to exchange information with 3D-CNNs with a cross-attention mechanism.

Video Recognition

SVTR: Scene Text Recognition with a Single Visual Model

4 code implementations30 Apr 2022 Yongkun Du, Zhineng Chen, Caiyan Jia, Xiaoting Yin, Tianlun Zheng, Chenxia Li, Yuning Du, Yu-Gang Jiang

Dominant scene text recognition models commonly contain two building blocks, a visual model for feature extraction and a sequence model for text transcription.

Scene Text Recognition

Deeper Insights into the Robustness of ViTs towards Common Corruptions

no code implementations26 Apr 2022 Rui Tian, Zuxuan Wu, Qi Dai, Han Hu, Yu-Gang Jiang

With Vision Transformers (ViTs) making great advances in a variety of computer vision tasks, recent literature have proposed various variants of vanilla ViTs to achieve better efficiency and efficacy.

Benchmarking Data Augmentation

Adaptive Split-Fusion Transformer

1 code implementation26 Apr 2022 Zixuan Su, Hao Zhang, Jingjing Chen, Lei Pang, Chong-Wah Ngo, Yu-Gang Jiang

Neural networks for visual content understanding have recently evolved from convolutional ones (CNNs) to transformers.

Image Classification

Wave-SAN: Wavelet based Style Augmentation Network for Cross-Domain Few-Shot Learning

1 code implementation15 Mar 2022 Yuqian Fu, Yu Xie, Yanwei Fu, Jingjing Chen, Yu-Gang Jiang

The key challenge of CD-FSL lies in the huge data shift between source and target domains, which is typically in the form of totally different visual styles.

cross-domain few-shot learning Self-Supervised Learning

Suspected Object Matters: Rethinking Model's Prediction for One-stage Visual Grounding

no code implementations10 Mar 2022 Yang Jiao, Zequn Jie, Jingjing Chen, Lin Ma, Yu-Gang Jiang

Recently, one-stage visual grounders attract high attention due to their comparable accuracy but significantly higher efficiency than two-stage grounders.

Object Visual Grounding

MORE: Multi-Order RElation Mining for Dense Captioning in 3D Scenes

1 code implementation10 Mar 2022 Yang Jiao, Shaoxiang Chen, Zequn Jie, Jingjing Chen, Lin Ma, Yu-Gang Jiang

3D dense captioning is a recently-proposed novel task, where point clouds contain more geometric information than the 2D counterpart.

3D dense captioning Dense Captioning +4

Cross-Modal Transferable Adversarial Attacks from Images to Videos

no code implementations CVPR 2022 Zhipeng Wei, Jingjing Chen, Zuxuan Wu, Yu-Gang Jiang

This paper investigates the transferability of adversarial perturbation across different modalities, i. e., leveraging adversarial perturbation generated on white-box image models to attack black-box video models.

Video Recognition

Unified Multimodal Pre-training and Prompt-based Tuning for Vision-Language Understanding and Generation

no code implementations10 Dec 2021 Tianyi Liu, Zuxuan Wu, Wenhan Xiong, Jingjing Chen, Yu-Gang Jiang

Our experiments show that there is a trade-off between understanding tasks and generation tasks while using the same model, and a feasible way to improve both tasks is to use more data.

Image-text matching Image-text Retrieval +9

BEVT: BERT Pretraining of Video Transformers

1 code implementation CVPR 2022 Rui Wang, Dongdong Chen, Zuxuan Wu, Yinpeng Chen, Xiyang Dai, Mengchen Liu, Yu-Gang Jiang, Luowei Zhou, Lu Yuan

This design is motivated by two observations: 1) transformers learned on image datasets provide decent spatial priors that can ease the learning of video transformers, which are often times computationally-intensive if trained from scratch; 2) discriminative clues, i. e., spatial and temporal information, needed to make correct predictions vary among different videos due to large intra-class and inter-class variations.

Action Recognition Representation Learning

AdaViT: Adaptive Vision Transformers for Efficient Image Recognition

1 code implementation CVPR 2022 Lingchen Meng, Hengduo Li, Bor-Chun Chen, Shiyi Lan, Zuxuan Wu, Yu-Gang Jiang, Ser-Nam Lim

To this end, we introduce AdaViT, an adaptive computation framework that learns to derive usage policies on which patches, self-attention heads and transformer blocks to use throughout the backbone on a per-input basis, aiming to improve inference efficiency of vision transformers with a minimal drop of accuracy for image recognition.

Efficient Video Transformers with Spatial-Temporal Token Selection

1 code implementation23 Nov 2021 Junke Wang, Xitong Yang, Hengduo Li, Li Liu, Zuxuan Wu, Yu-Gang Jiang

Video transformers have achieved impressive results on major video recognition benchmarks, which however suffer from high computational cost.

Video Recognition

CDistNet: Perceiving Multi-Domain Character Distance for Robust Text Recognition

3 code implementations22 Nov 2021 Tianlun Zheng, Zhineng Chen, Shancheng Fang, Hongtao Xie, Yu-Gang Jiang

In this paper, we propose a novel module called Multi-Domain Character Distance Perception (MDCDP) to establish a visually and semantically related position embedding.

Position Scene Text Recognition

Semi-Supervised Vision Transformers

1 code implementation22 Nov 2021 Zejia Weng, Xitong Yang, Ang Li, Zuxuan Wu, Yu-Gang Jiang

Surprisingly, we show Vision Transformers perform significantly worse than Convolutional Neural Networks when only a small set of labeled data is available.

image-classification Inductive Bias +1

Attacking Video Recognition Models with Bullet-Screen Comments

1 code implementation29 Oct 2021 Kai Chen, Zhipeng Wei, Jingjing Chen, Zuxuan Wu, Yu-Gang Jiang

On both UCF-101 and HMDB-51 datasets, our BSC attack method can achieve about 90\% fooling rate when attacking three mainstream video recognition models, while only occluding \textless 8\% areas in the video.

Adversarial Attack Adversarial Attack on Video Classification +2

Boosting the Transferability of Video Adversarial Examples via Temporal Translation

1 code implementation18 Oct 2021 Zhipeng Wei, Jingjing Chen, Zuxuan Wu, Yu-Gang Jiang

To this end, we propose to boost the transferability of video adversarial examples for black-box attacks on video recognition models.

Adversarial Attack Translation +1

Two-stage Visual Cues Enhancement Network for Referring Image Segmentation

1 code implementation9 Oct 2021 Yang Jiao, Zequn Jie, Weixin Luo, Jingjing Chen, Yu-Gang Jiang, Xiaolin Wei, Lin Ma

Referring Image Segmentation (RIS) aims at segmenting the target object from an image referred by one given natural language expression.

Image Segmentation Retrieval +2

Self-supervised Learning for Semi-supervised Temporal Language Grounding

no code implementations23 Sep 2021 Fan Luo, Shaoxiang Chen, Jingjing Chen, Zuxuan Wu, Yu-Gang Jiang

Given a text description, Temporal Language Grounding (TLG) aims to localize temporal boundaries of the segments that contain the specified semantics in an untrimmed video.

Contrastive Learning Pseudo Label +2

Towards Transferable Adversarial Attacks on Vision Transformers

2 code implementations9 Sep 2021 Zhipeng Wei, Jingjing Chen, Micah Goldblum, Zuxuan Wu, Tom Goldstein, Yu-Gang Jiang

We evaluate the transferability of attacks on state-of-the-art ViTs, CNNs and robustly trained CNNs.

A Multimodal Framework for Video Ads Understanding

no code implementations29 Aug 2021 Zejia Weng, Lingchen Meng, Rui Wang, Zuxuan Wu, Yu-Gang Jiang

There is a growing trend in placing video advertisements on social platforms for online marketing, which demands automatic approaches to understand the contents of advertisements effectively.

Marketing Optical Character Recognition +5

Revisiting Adversarial Robustness Distillation: Robust Soft Labels Make Student Better

1 code implementation ICCV 2021 Bojia Zi, Shihao Zhao, Xingjun Ma, Yu-Gang Jiang

We empirically demonstrate the effectiveness of our RSLAD approach over existing adversarial training and distillation methods in improving the robustness of small models against state-of-the-art attacks including the AutoAttack.

Adversarial Robustness Knowledge Distillation

FT-TDR: Frequency-guided Transformer and Top-Down Refinement Network for Blind Face Inpainting

no code implementations10 Aug 2021 Junke Wang, Shaoxiang Chen, Zuxuan Wu, Yu-Gang Jiang

Blind face inpainting refers to the task of reconstructing visual contents without explicitly indicating the corrupted regions in a face image.

Facial Inpainting

Meta-FDMixup: Cross-Domain Few-Shot Learning Guided by Labeled Target Data

1 code implementation26 Jul 2021 Yuqian Fu, Yanwei Fu, Yu-Gang Jiang

Secondly, a novel disentangle module together with a domain classifier is proposed to extract the disentangled domain-irrelevant and domain-specific features.

cross-domain few-shot learning

Can Action be Imitated? Learn to Reconstruct and Transfer Human Dynamics from Videos

no code implementations25 Jul 2021 Yuqian Fu, Yanwei Fu, Yu-Gang Jiang

To achieve this, a novel Mesh-based Video Action Imitation (M-VAI) method is proposed by us.

Human Dynamics

Towards Bridging Event Captioner and Sentence Localizer for Weakly Supervised Dense Event Captioning

no code implementations CVPR 2021 Shaoxiang Chen, Yu-Gang Jiang

Dense Event Captioning (DEC) aims to jointly localize and describe multiple events of interest in untrimmed videos, which is an advancement of the conventional video captioning task (generating a single sentence description for a trimmed video).

Sentence Video Captioning

Cross-domain Contrastive Learning for Unsupervised Domain Adaptation

1 code implementation10 Jun 2021 Rui Wang, Zuxuan Wu, Zejia Weng, Jingjing Chen, Guo-Jun Qi, Yu-Gang Jiang

Unsupervised domain adaptation (UDA) aims to transfer knowledge learned from a fully-labeled source domain to a different unlabeled target domain.

Clustering Contrastive Learning +4

VideoLT: Large-scale Long-tailed Video Recognition

1 code implementation ICCV 2021 Xing Zhang, Zuxuan Wu, Zejia Weng, Huazhu Fu, Jingjing Chen, Yu-Gang Jiang, Larry Davis

In this paper, we introduce VideoLT, a large-scale long-tailed video recognition dataset, as a step toward real-world video recognition.

image-classification Image Classification +1

M2TR: Multi-modal Multi-scale Transformers for Deepfake Detection

3 code implementations20 Apr 2021 Junke Wang, Zuxuan Wu, Wenhao Ouyang, Xintong Han, Jingjing Chen, Ser-Nam Lim, Yu-Gang Jiang

The widespread dissemination of Deepfakes demands effective approaches that can detect perceptually convincing forged images.

DeepFake Detection Face Swapping +1

What Do Deep Nets Learn? Class-wise Patterns Revealed in the Input Space

no code implementations18 Jan 2021 Shihao Zhao, Xingjun Ma, Yisen Wang, James Bailey, Bo Li, Yu-Gang Jiang

In this paper, we focus on image classification and propose a method to visualize and understand the class-wise knowledge (patterns) learned by DNNs under three different settings including natural, backdoor and adversarial.

image-classification Image Classification

WildDeepfake: A Challenging Real-World Dataset for Deepfake Detection

1 code implementation5 Jan 2021 Bojia Zi, Minghao Chang, Jingjing Chen, Xingjun Ma, Yu-Gang Jiang

WildDeepfake is a small dataset that can be used, in addition to existing datasets, to develop and test the effectiveness of deepfake detectors against real-world deepfakes.

DeepFake Detection Face Swapping

Motion Guided Region Message Passing for Video Captioning

no code implementations ICCV 2021 Shaoxiang Chen, Yu-Gang Jiang

In this paper, we aim at designing a spatial information extraction and aggregation method for video captioning without the need of external object detectors.

Decoder Video Captioning

Colonoscopy Polyp Detection: Domain Adaptation From Medical Report Images to Real-time Videos

no code implementations31 Dec 2020 Zhi-Qin Zhan, Huazhu Fu, Yan-Yao Yang, Jingjing Chen, Jie Liu, Yu-Gang Jiang

However, there are several issues between the image-based training and video-based inference, including domain differences, lack of positive samples, and temporal smoothness.

Domain Adaptation

Depth Guided Adaptive Meta-Fusion Network for Few-shot Video Recognition

1 code implementation20 Oct 2020 Yuqian Fu, Li Zhang, Junke Wang, Yanwei Fu, Yu-Gang Jiang

Humans can easily recognize actions with only a few examples given, while the existing video recognition models still heavily rely on the large-scale labeled data inputs.

Few Shot Action Recognition Meta-Learning +2

Imbalanced Gradients: A New Cause of Overestimated Adversarial Robustness

no code implementations28 Sep 2020 Linxi Jiang, Xingjun Ma, Zejia Weng, James Bailey, Yu-Gang Jiang

Evaluating the robustness of a defense model is a challenging task in adversarial robustness research.

Adversarial Robustness

Multi-modal Cooking Workflow Construction for Food Recipes

no code implementations20 Aug 2020 Liangming Pan, Jingjing Chen, Jianlong Wu, Shaoteng Liu, Chong-Wah Ngo, Min-Yen Kan, Yu-Gang Jiang, Tat-Seng Chua

Understanding food recipe requires anticipating the implicit causal effects of cooking actions, such that the recipe can be converted into a graph describing the temporal workflow of the recipe.

Common Sense Reasoning Decoder

Learning Modality Interaction for Temporal Sentence Localization and Event Captioning in Videos

no code implementations ECCV 2020 Shaoxiang Chen, Wenhao Jiang, Wei Liu, Yu-Gang Jiang

Inspired by the fact that there exist cross-modal interactions in the human brain, we propose a novel method for learning pairwise modality interactions in order to better exploit complementary information for each pair of modalities in videos and thus improve performances on both tasks.

Sentence

Imbalanced Gradients: A Subtle Cause of Overestimated Adversarial Robustness

1 code implementation24 Jun 2020 Xingjun Ma, Linxi Jiang, Hanxun Huang, Zejia Weng, James Bailey, Yu-Gang Jiang

Evaluating the robustness of a defense model is a challenging task in adversarial robustness research.

Adversarial Robustness

Long-Term Cloth-Changing Person Re-identification

no code implementations26 May 2020 Xuelin Qian, Wenxuan Wang, Li Zhang, Fangrui Zhu, Yanwei Fu, Tao Xiang, Yu-Gang Jiang, xiangyang xue

Specifically, we consider that under cloth-changes, soft-biometrics such as body shape would be more reliable.

Cloth-Changing Person Re-Identification

Sketch-BERT: Learning Sketch Bidirectional Encoder Representation from Transformers by Self-supervised Learning of Sketch Gestalt

1 code implementation CVPR 2020 Hangyu Lin, Yanwei Fu, Yu-Gang Jiang, xiangyang xue

Unfortunately, the representation learned by SketchRNN is primarily for the generation tasks, rather than the other tasks of recognition and retrieval of sketches.

Retrieval Self-Supervised Learning +1

Clean-Label Backdoor Attacks on Video Recognition Models

1 code implementation CVPR 2020 Shihao Zhao, Xingjun Ma, Xiang Zheng, James Bailey, Jingjing Chen, Yu-Gang Jiang

We propose the use of a universal adversarial trigger as the backdoor trigger to attack video recognition models, a situation where backdoor attacks are likely to be challenged by the above 4 strict conditions.

Backdoor Attack backdoor defense +3

Learning to Augment Expressions for Few-shot Fine-grained Facial Expression Recognition

no code implementations17 Jan 2020 Wenxuan Wang, Yanwei Fu, Qiang Sun, Tao Chen, Chenjie Cao, Ziqi Zheng, Guoqiang Xu, Han Qiu, Yu-Gang Jiang, xiangyang xue

Considering the phenomenon of uneven data distribution and lack of samples is common in real-world scenarios, we further evaluate several tasks of few-shot expression learning by virtue of our F2ED, which are to recognize the facial expressions given only few training instances.

Facial Expression Recognition Facial Expression Recognition (FER) +1

LiteEval: A Coarse-to-Fine Framework for Resource Efficient Video Recognition

no code implementations NeurIPS 2019 Zuxuan Wu, Caiming Xiong, Yu-Gang Jiang, Larry S. Davis

This paper presents LiteEval, a simple yet effective coarse-to-fine framework for resource efficient video recognition, suitable for both online and offline scenarios.

Video Recognition

Heuristic Black-box Adversarial Attacks on Video Recognition Models

1 code implementation21 Nov 2019 Zhipeng Wei, Jingjing Chen, Xingxing Wei, Linxi Jiang, Tat-Seng Chua, Fengfeng Zhou, Yu-Gang Jiang

To overcome this challenge, we propose a heuristic black-box attack model that generates adversarial perturbations only on the selected frames and regions.

Adversarial Attack Video Recognition

DeepEnFM: Deep neural networks with Encoder enhanced Factorization Machine

no code implementations25 Sep 2019 Qiang Sun, Zhinan Cheng, Yanwei Fu, Wenxuan Wang, Yu-Gang Jiang, xiangyang xue

Instead of learning the cross features directly, DeepEnFM adopts the Transformer encoder as a backbone to align the feature embeddings with the clues of other fields.

Click-Through Rate Prediction

Black-box Adversarial Attacks on Video Recognition Models

no code implementations10 Apr 2019 Linxi Jiang, Xingjun Ma, Shaoxiang Chen, James Bailey, Yu-Gang Jiang

Using three benchmark video datasets, we demonstrate that V-BAD can craft both untargeted and targeted attacks to fool two state-of-the-art deep video recognition models.

Video Recognition

Instance-level Sketch-based Retrieval by Deep Triplet Classification Siamese Network

no code implementations28 Nov 2018 Peng Lu, Hangyu Lin, Yanwei Fu, Shaogang Gong, Yu-Gang Jiang, xiangyang xue

Additionally, to study the tasks of sketch-based hairstyle retrieval, this paper contributes a new instance-level photo-sketch dataset - Hairstyle Photo-Sketch dataset, which is composed of 3600 sketches and photos, and 2400 sketch-photo pairs.

General Classification Retrieval +3

Composite Binary Decomposition Networks

no code implementations16 Nov 2018 You Qiaoben, Zheng Wang, Jianguo Li, Yinpeng Dong, Yu-Gang Jiang, Jun Zhu

Binary neural networks have great resource and computing efficiency, while suffer from long training procedure and non-negligible accuracy drops, when comparing to the full-precision counterparts.

General Classification image-classification +4

Non-local NetVLAD Encoding for Video Classification

no code implementations29 Sep 2018 Yongyi Tang, Xing Zhang, Jingwen Wang, Shaoxiang Chen, Lin Ma, Yu-Gang Jiang

This paper describes our solution for the 2$^\text{nd}$ YouTube-8M video understanding challenge organized by Google AI.

Classification General Classification +3

Object Detection from Scratch with Deep Supervision

1 code implementation25 Sep 2018 Zhiqiang Shen, Zhuang Liu, Jianguo Li, Yu-Gang Jiang, Yurong Chen, xiangyang xue

Thus, a better solution to handle these critical problems is to train object detectors from scratch, which motivates our proposed method.

General Classification Object +2

Recurrent Fusion Network for Image Captioning

no code implementations ECCV 2018 Wenhao Jiang, Lin Ma, Yu-Gang Jiang, Wei Liu, Tong Zhang

In this paper, in order to exploit the complementary information from multiple encoders, we propose a novel Recurrent Fusion Network (RFNet) for tackling image captioning.

Decoder Image Captioning

Unsupervised Image-to-Image Translation with Stacked Cycle-Consistent Adversarial Networks

no code implementations ECCV 2018 Minjun Li, Hao-Zhi Huang, Lin Ma, Wei Liu, Tong Zhang, Yu-Gang Jiang

Recent studies on unsupervised image-to-image translation have made a remarkable progress by training a pair of generative adversarial networks with a cycle-consistent loss.

Translation Unsupervised Image-To-Image Translation

Multi-level Semantic Feature Augmentation for One-shot Learning

1 code implementation15 Apr 2018 Zitian Chen, Yanwei Fu, yinda zhang, Yu-Gang Jiang, xiangyang xue, Leonid Sigal

In semantic space, we search for related concepts, which are then projected back into the image feature spaces by the decoder portion of the TriNet.

Decoder Novel Concepts +1

Social Anchor-Unit Graph Regularized Tensor Completion for Large-Scale Image Retagging

no code implementations12 Apr 2018 Jinhui Tang, Xiangbo Shu, Zechao Li, Yu-Gang Jiang, Qi Tian

Recent approaches simultaneously explore visual, user and tag information to improve the performance of image retagging by constructing and exploring an image-tag-user graph.

Graph Learning TAG

Pixel2Mesh: Generating 3D Mesh Models from Single RGB Images

5 code implementations ECCV 2018 Nanyang Wang, yinda zhang, Zhuwen Li, Yanwei Fu, Wei Liu, Yu-Gang Jiang

We propose an end-to-end deep learning architecture that produces a 3D shape in triangular mesh from a single color image.

3D geometry 3D Object Reconstruction

Learning to score the figure skating sports videos

1 code implementation8 Feb 2018 Chengming Xu, Yanwei Fu, Bing Zhang, Zitian Chen, Yu-Gang Jiang, xiangyang xue

This paper targets at learning to score the figure skating sports videos.

Cannot find the paper you are looking for? You can Submit a new open access paper.