Search Results for author:
Found 27 papers, 9 papers with code
De-confounded Data-free Knowledge Distillation for Handling Distribution Shifts
Data-Free Knowledge Distillation (DFKD) is a promising task to train high-performance small models to enhance actual deployment without relying on the original training data.
Improving Adversarial Transferability of Visual-Language Pre-training Models through Collaborative Multimodal Interaction
Existing work rarely studies the transferability of attacks on VLP models, resulting in a substantial performance gap from white-box attacks.
OneTracker: Unifying Visual Object Tracking with Foundation Models and Efficient Tuning
To evaluate the effectiveness of our general framework OneTracker, which is consisted of Foundation Tracker and Prompt Tracker, we conduct extensive experiments on 6 popular tracking tasks across 11 benchmarks and our OneTracker outperforms other models and achieves state-of-the-art performance.
ClickVOS: Click Video Object Segmentation
To address these limitations, we propose the setting named Click Video Object Segmentation (ClickVOS) which segments objects of interest across the whole video according to a single click per object in the first frame.
Towards Multimodal Human Intention Understanding Debiasing via Subject-Deconfounding
Multimodal intention understanding (MIU) is an indispensable component of human expression analysis (e. g., sentiment or humor) from heterogeneous modalities, including visual postures, linguistic contents, and acoustic behaviors.
Towards Multimodal Sentiment Analysis Debiasing via Bias Purification
In the inference phase, given a factual multimodal input, MCIS imagines two counterfactual scenarios to purify and mitigate these biases.
Delving into Decision-based Black-box Attacks on Semantic Segmentation
We conduct adversarial robustness evaluation on 5 models from Cityscapes and ADE20K under 8 attacks.
Exploring Decision-based Black-box Attacks on Face Forgery Detection
Further, the fake faces by our method can pass face forgery detection and face recognition, which exposes the security problems of face forgery detectors.
Improving Generalization in Visual Reinforcement Learning via Conflict-aware Gradient Agreement Augmentation
In particular, CG2A develops a Gradient Agreement Solver to adaptively balance the varying gradient magnitudes, and introduces a Soft Gradient Surgery strategy to alleviate the gradient conflicts.
Sampling to Distill: Knowledge Transfer from Open-World Data
Then, we introduce a low-noise representation to alleviate the domain shifts and build a structured relationship of multiple data examples to exploit data knowledge.
AIDE: A Vision-Driven Multi-View, Multi-Modal, Multi-Tasking Dataset for Assistive Driving Perception
Driver distraction has become a significant cause of severe traffic accidents over the past decade.
Query-Efficient Decision-based Black-Box Patch Attack
In this work, we first explore the decision-based patch attack.
OpenVIS: Open-vocabulary Video Instance Segmentation
Open-vocabulary Video Instance Segmentation (OpenVIS) can simultaneously detect, segment, and track arbitrary object categories in a video, without being constrained to categories seen during training.
Non-rigid Point Cloud Registration for Middle Ear Diagnostics with Endoscopic Optical Coherence Tomography
To overcome the lack of labeled training data, a fast and effective generation pipeline in Blender3D is designed to simulate middle ear shapes and extract in-vivo noisy and partial point clouds.
Out of Thin Air: Exploring Data-Free Adversarial Robustness Distillation
Adversarial Robustness Distillation (ARD) is a promising task to solve the issue of limited adversarial robustness of small capacity models while optimizing the expensive computational costs of Adversarial Training (AT).
Context De-confounded Emotion Recognition
However, a long-overlooked issue is that a context bias in existing datasets leads to a significantly unbalanced distribution of emotional states among different context scenarios.
Efficient Decision-based Black-box Patch Attacks on Video Recognition
First, STDE introduces target videos as patch textures and only adds patches on keyframes that are adaptively selected by temporal difference.
Explicit and Implicit Knowledge Distillation via Unlabeled Data
Data-free knowledge distillation is a challenging model lightweight task for scenarios in which the original dataset is not available.
Adversarial Contrastive Distillation with Adaptive Denoising
To this end, we propose a novel structured ARD method called Contrastive Relationship DeNoise Distillation (CRDND).
Boosting the Transferability of Adversarial Attacks with Global Momentum Initialization
To tackle these issues, we propose Global Momentum Initialization (GI) to suppress gradient elimination and help search for the global optimum.
LVOS: A Benchmark for Long-term Video Object Segmentation
The videos in our LVOS last 1. 59 minutes on average, which is 20 times longer than videos in existing VOS datasets.
Shape Matters: Deformable Patch Attack
Though deep neural networks (DNNs) have demonstrated excellent performance in computer vision, they are susceptible and vulnerable to carefully crafted adversarial examples which can mislead DNNs to incorrect outputs.
DPCNet: Dual Path Multi-Excitation Collaborative Network for Facial Expression Representation Learning in Videos
To obtain consistent prediction probabilities from the dual path, we further propose a dual path regularization loss, aiming to minimize the divergence between the distributions of two-path embeddings.
Ranked #13 on
Dynamic Facial Expression Recognition
on DFEW
Dynamic Facial Expression Recognition
Representation Learning
Towards Practical Certifiable Patch Defense with Vision Transformer
To move towards a practical certifiable patch defense, we introduce Vision Transformer (ViT) into the framework of Derandomized Smoothing (DS).
Efficient universal shuffle attack for visual object tracking
Recently, adversarial attacks have been applied in visual object tracking to deceive deep trackers by injecting imperceptible perturbations into video frames.
CMUA-Watermark: A Cross-Model Universal Adversarial Watermark for Combating Deepfakes
Then, we design a two-level perturbation fusion strategy to alleviate the conflict between the adversarial watermarks generated by different facial images and models.
RPATTACK: Refined Patch Attack on General Object Detectors
Firstly, we propose a patch selection and refining scheme to find the pixels which have the greatest importance for attack and remove the inconsequential perturbations gradually.
