Search Results for author:
Found 58 papers, 40 papers with code
VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training
Pre-training video transformers on extra large-scale datasets is generally required to achieve premier performance on relatively small datasets.
Ranked #5 on
Self-Supervised Action Recognition
on HMDB51
Efficient Video Action Detection with Token Dropout and Context Refinement
Our EVAD consists of two specialized designs for video action detection.
Parser-Free Virtual Try-on via Distilling Appearance Flows
A recent pioneering work employed knowledge distillation to reduce the dependency of human parsing, where the try-on images produced by a parser-based method are used as supervisions to train a "student" network without relying on segmentation, making the student mimic the try-on ability of the parser-based model.
Ranked #1 on
Virtual Try-on
on MPV
Rethinking Image Inpainting via a Mutual Encoder-Decoder with Feature Equalizations
We use CNN features from the deep and shallow layers of the encoder to represent structures and textures of an input image, respectively.
AdaptFormer: Adapting Vision Transformers for Scalable Visual Recognition
To address this challenge, we propose an effective adaptation approach for Transformer, namely AdaptFormer, which can adapt the pre-trained ViTs into many different image and video tasks efficiently.
DynaMixer: A Vision MLP Architecture with Dynamic Mixing
In contrast with vision transformers and CNNs, the success of MLP-like models shows that simple information fusion operations among tokens and channels can yield a good representation power for deep recognition models.
Unsupervised Deep Representation Learning for Real-Time Tracking
The advancement of visual tracking has continuously been brought by deep learning models.
MVF-Net: Multi-View 3D Face Morphable Model Regression
The main ingredient of the view alignment loss is a differentiable dense optical flow estimator that can backpropagate the alignment errors between an input view and a synthetic rendering from another input view, which is projected to the target view through the 3D shape to be inferred.
EViT: Expediting Vision Transformers via Token Reorganizations
Second, by maintaining the same computational cost, our method empowers ViTs to take more image tokens as input for recognition accuracy improvement, where the image tokens are from higher resolution images.
Not All Patches are What You Need: Expediting Vision Transformers via Token Reorganizations
Second, by maintaining the same computational cost, our method empowers ViTs to take more image tokens as input for recognition accuracy improvement, where the image tokens are from higher resolution images.
Ranked #4 on
Efficient ViTs
on ImageNet-1K (with DeiT-S)
ArtFlow: Unbiased Image Style Transfer via Reversible Neural Flows
The forward inference projects input images into deep features, while the backward inference remaps deep features back to input images in a lossless and unbiased way.
VideoMoCo: Contrastive Video Representation Learning with Temporally Adversarial Examples
By empowering the temporal robustness of the encoder and modeling the temporal decay of the keys, our VideoMoCo improves MoCo temporally based on contrastive learning.
Ranked #76 on
Action Recognition
on HMDB-51
PD-GAN: Probabilistic Diverse GAN for Image Inpainting
To this end, we propose spatially probabilistic diversity normalization (SPDNorm) inside the modulation to model the probability of generating a pixel conditioned on the context information.
Self-supervised Learning of Adversarial Example: Towards Good Generalizations for Deepfake Detection
Following this principle, we propose to enrich the "diversity" of forgeries by synthesizing augmented forgeries with a pool of forgery configurations and strengthen the "sensitivity" to the forgeries by enforcing the model to predict the forgery configurations.
Revitalizing CNN Attentions via Transformers in Self-Supervised Visual Representation Learning
Motivated by the transformers that explore visual attention effectively in recognition scenarios, we propose a CNN Attention REvitalization (CARE) framework to train attentive CNN encoders guided by transformers in SSL.
Revitalizing CNN Attention via Transformers in Self-Supervised Visual Representation Learning
Motivated by the transformers that explore visual attention effectively in recognition scenarios, we propose a CNN Attention REvitalization (CARE) framework to train attentive CNN encoders guided by transformers in SSL.
Disentangled Cycle Consistency for Highly-realistic Virtual Try-On
To this end, DCTON can be naturally trained in a self-supervised manner following cycle consistency learning.
Advancing Vision Transformers with Group-Mix Attention
The attention map is computed based on the mixtures of tokens and group proxies and used to re-combine the tokens and groups in Value.
Bridging Vision and Language Encoders: Parameter-Efficient Tuning for Referring Image Segmentation
Parameter Efficient Tuning (PET) has gained attention for reducing the number of parameters while maintaining performance and providing better hardware resource savings, but few studies investigate dense prediction tasks and interaction between modalities.
Ranked #2 on
Referring Expression Segmentation
on RefCOCO
Self-supervised Learning of Detailed 3D Face Reconstruction
The displacement map and the coarse model are used to render a final detailed face, which again can be compared with the original input image to serve as a photometric loss for the second stage.
Image Inpainting via Iteratively Decoupled Probabilistic Modeling
To achieve high-quality results with low computational cost, we present a novel pixel spread model (PSM) that iteratively employs decoupled probabilistic modeling, combining the optimization efficiency of GANs with the prediction tractability of probabilistic models.
DeFLOCNet: Deep Image Editing via Flexible Low-level Controls
While existing methods combine an input image and these low-level controls for CNN inputs, the corresponding feature representations are not sufficient to convey user intentions, leading to unfaithfully generated content.
IoU Attack: Towards Temporally Coherent Black-Box Adversarial Attack for Visual Object Tracking
Recently, adversarial attack has been applied to visual object tracking to evaluate the robustness of deep trackers.
Dynamic Scene Deblurring Using Spatially Variant Recurrent Neural Networks
The proposed network is composed of three deep convolutional neural networks (CNNs) and a recurrent neural network (RNN).
Ranked #10 on
Deblurring
on RealBlur-R (trained on GoPro)
(SSIM (sRGB) metric)
One Model to Edit Them All: Free-Form Text-Driven Image Manipulation with Semantic Modulations
Based on the visual latent space of StyleGAN[21] and text embedding space of CLIP[34], studies focus on how to map these two latent spaces for text-driven attribute manipulations.
Delving StyleGAN Inversion for Image Editing: A Foundation Latent Space Viewpoint
In this work, we propose to first obtain the precise latent code in foundation latent space $\mathcal{W}$.
Human MotionFormer: Transferring Human Motions with Vision Transformers
In this paper, we propose Human MotionFormer, a hierarchical ViT framework that leverages global and local perceptions to capture large and subtle motion matching, respectively.
Robust Tracking against Adversarial Attacks
On one hand, we add the temporal perturbations into the original video sequences as adversarial examples to greatly degrade the tracking performance.
Improved Test-Time Adaptation for Domain Generalization
Generally, a TTT strategy hinges its performance on two main factors: selecting an appropriate auxiliary TTT task for updating and identifying reliable parameters to update during the test phase.
InstructDET: Diversifying Referring Object Detection with Generalized Instructions
In order to encompass common detection expressions, we involve emerging vision-language model (VLM) and large language model (LLM) to generate instructions guided by text prompts and object bbxs, as the generalizations of foundation models are effective to produce human-like expressions (e. g., describing object property, category, and relationship).
Dynamic Tuning Towards Parameter and Inference Efficiency for ViT Adaptation
Existing parameter-efficient fine-tuning (PEFT) methods have achieved significant success on vision transformers (ViTs) adaptation by improving parameter efficiency.
Advancing Visual Grounding with Scene Knowledge: Benchmark and Method
Therefore, in this paper, we propose a novel benchmark of \underline{S}cene \underline{K}nowledge-guided \underline{V}isual \underline{G}rounding (SK-VG), where the image content and referring expressions are not sufficient to ground the target objects, forcing the models to have a reasoning ability on the long-form scene knowledge.
Domain Generalization via Rationale Invariance
Specifically, we propose treating the element-wise contributions to the final results as the rationale for making a decision and representing the rationale for each sample as a matrix.
Rethinking Image Deraining via Rain Streaks and Vapors
Single image deraining regards an input image as a fusion of a background image, a transmission map, rain streaks, and atmosphere light.
Real-Time Correlation Tracking via Joint Model Compression and Transfer
In the distillation process, we propose a fidelity loss to enable the student network to maintain the representation capability of the teacher network.
TransZero++: Cross Attribute-Guided Transformer for Zero-Shot Learning
Analogously, VAT uses the similar feature augmentation encoder to refine the visual features, which are further applied in visual$\rightarrow$attribute decoder to learn visual-based attribute features.
Stabilized Medical Image Attacks
However, a threat to these systems arises that adversarial attacks make CNNs vulnerable.
A Causal Inspired Early-Branching Structure for Domain Generalization
By d-separation, we observe that the causal feature can be further characterized by being independent of the domain conditioned on the object, and we propose the following two strategies as complements for the basic framework.
Image Correction via Deep Reciprocating HDR Transformation
Given an input LDR image, we first reconstruct the missing details in the HDR domain.
VITAL: VIsual Tracking via Adversarial Learning
To augment positive samples, we use a generative network to randomly generate masks, which are applied to adaptively dropout input features to capture a variety of appearance changes.
Stylizing Face Images via Multiple Exemplars
We address the problem of transferring the style of a headshot photo to face images.
CREST: Convolutional Residual Learning for Visual Tracking
Our method integrates feature extraction, response map generation as well as model update into the neural networks for an end-to-end training.
Fast Preprocessing for Robust Face Sketch Synthesis
Exemplar-based face sketch synthesis methods usually meet the challenging problem that input photos are captured in different lighting conditions from training photos.
Learning to Hallucinate Face Images via Component Generation and Enhancement
We propose a two-stage method for face hallucination.
Deformable Object Tracking with Gated Fusion
In addition, we also propose a gated fusion scheme to control how the variations captured by the deformable convolution affect the original appearance.
Deep Attentive Tracking via Reciprocative Learning
Visual attention, derived from cognitive neuroscience, facilitates human perception on the most pertinent subset of the sensory data.
Joint Face Hallucination and Deblurring via Structure Generation and Detail Enhancement
We first propose a facial component guided deep Convolutional Neural Network (CNN) to restore a coarse face image, which is denoted as the base image where the facial component is automatically generated from the input face image.
Look Deeper into Depth: Monocular Depth Estimation with Semantic Booster and Attention-Driven Loss
Monocular depth estimation benefits greatly from learning based techniques.
Stabilized Medical Attacks
We further analyze the KL-divergence of the proposed loss function and find that the loss stabilization term makes the perturbations updated towards a fixed objective spot while deviating from the ground truth.
MetaDance: Few-shot Dancing Video Retargeting via Temporal-aware Meta-learning
Dancing video retargeting aims to synthesize a video that transfers the dance movements from a source video to a target person.
CycleACR: Cycle Modeling of Actor-Context Relations for Video Action Detection
Existing studies model each actor and scene relation to improve action recognition.
Soft Neighbors are Positive Supporters in Contrastive Visual Representation Learning
We evaluate our soft neighbor contrastive learning method (SNCLR) on standard visual recognition benchmarks, including image classification, object detection, and instance segmentation.
Evolving Semantic Prototype Improves Generative Zero-Shot Learning
After alignment, synthesized sample features from unseen classes are closer to the real sample features and benefit DSP to improve existing generative ZSL methods by 8. 5\%, 8. 0\%, and 9. 7\% on the standard CUB, SUN AWA2 datasets, the significant performance improvement indicates that evolving semantic prototype explores a virgin field in ZSL.
Speed Co-Augmentation for Unsupervised Audio-Visual Pre-training
This work aims to improve unsupervised audio-visual pre-training.
Both Diverse and Realism Matter: Physical Attribute and Style Alignment for Rainy Image Generation
To solve this dilemma, we propose a physical alignment and controllable generation network (PCGNet) for diverse and realistic rain generation.
HeadArtist: Text-conditioned 3D Head Generation with Self Score Distillation
The noisy image, landmarks, and text condition are then fed into the frozen ControlNet twice for noise prediction.
