Accv 2022

186 papers

Top Research Directions

Research directions most represented by Accv 2022 papers.

Cross Task Generalization in Image Restoration 3
Plug and Play Attention Modules for Deep Networks 2
Composite Loss Functions for Perceptual Image Restoration 2
Hybrid Spatial and Channel Attention for Feature Refinement 2
Mitigating Error Propagation in Pseudo Labeling 2
Overcoming Metadata Fragility in Image Processing and Restoration 1
Real-Time Depth Inference for Autonomous Systems 1
Uncertainty Guided Spatial Refinement for Efficient Vision 1
Regularizing Representations with Self-Supervised Auxiliary Tasks 1
Feature Matching for Stable Generative Adversarial Training 1
Counterfactual Probing for Bias in Generative Models 1
Region Based Active Learning for Dense Prediction Tasks 1
One Stream Transformers for Visual Object Tracking 1
Constant Time Multi Person Pose Estimation 1
Image Restoration for Downstream Machine Perception 1

Top Authors

AuthorPapers
Haofeng Zhang4
Jin Xie3
Yanwei Fu3
Andrew Zisserman2
Bing Li2
Di Xie2
Fahad Shahbaz Khan2
Hao Huang2
Hongdong Li2
Huchuan Lu2
Liansheng Zhuang2
Lianwen Jin2
Minh N. Do2
Pascal Fua2
Piotr Koniusz2
Qingsen Yan2
Shiliang Pu2
Siyu Zhu2
Tingting Jiang2
Wanqing Li2

Impactful Papers

Papers from Accv 2022 that shaped research clusters.

Foundational

Influential

Papers

Title
Learning Using Privileged Information for Zero-Shot Action Recognition
Lightweight Alpha Matting Network Using Distillation-Based Channel Pruning
Lightweight Image Matting via Efficient Non-local Guidance
Looking from a Higher-Level Perspective: Attention and Recognition Enhanced Multi-scale Scene Text Segmentation
LSMD-Net: LiDAR-Stereo Fusion with Mixture Density Network for Depth Sensing
MatchFormer: Interleaving Attention in Transformers for Feature Matching
MaxGNR: A Dynamic Weight Strategy via Maximizing Gradient-to-Noise Ratio for Multi-task Learning
Meta-Det3D: Learn to Learn Few-Shot 3D Object Detection
Meta-prototype Decoupled Training for Long-Tailed Learning
MGTR: End-to-End Mutual Gaze Detection with Transformer
Multi-Branch Network with Ensemble Learning for Text Removal in the Wild
Multi-modal Characteristic Guided Depth Completion Network
Multi-modal Segment Assemblage Network for Ad Video Editing with Importance-Coherence Reward
Multi-scale Wavelet Transformer for Face Forgery Detection
Multi-stream Fusion for Class Incremental Learning in Pill Image Classification
Multiview Detection with Cardboard Human Modeling
MUSH: Multi-scale Hierarchical Feature Extraction for Semantic Image Synthesis
Neighborhood Region Smoothing Regularization for Finding Flat Minima in Deep Neural Networks
NEO-3DF: Novel Editing-Oriented 3D Face Creation and Reconstruction
Neural Deformable Voxel Grid for Fast Optimization of Dynamic View Synthesis
Neural Plenoptic Sampling: Learning Light-Field from Thousands of Imaginary Eyes
Neural Puppeteer: Keypoint-Based Neural Rendering of Dynamic Shapes
Neural Residual Flow Fields for Efficient Video Representations
NoiseTransfer: Image Noise Generation with Contrastive Embeddings
Not End-to-End: Explore Multi-Stage Architecture for Online Surgical Phase Recognition
Object Detection in Foggy Scenes by Embedding Depth and Reconstruction into Domain Adaptation
Occluded Facial Expression Recognition Using Self-supervised Learning
PathTR: Context-Aware Memory Transformer for Tumor Localization in Gigapixel Pathology Images
Physical Passive Patch Adversarial Attacks on Visual Odometry Systems
PixMamba: Leveraging State Space Models in a Dual-Level Architecture for Underwater Image Enhancement
PlainUSR: Chasing Faster ConvNet for Efficient Super-Resolution
Point Cloud Upsampling via Cascaded Refinement Network
PointFormer: A Dual Perception Attention-Based Network for Point Cloud Classification
PPR-Net: Patch-Based Multi-scale Pyramid Registration Network for Defect Detection of Printed Label
Progressive Attentional Manifold Alignment for Arbitrary Style Transfer
PS-ARM: An End-to-End Attention-Aware Relation Mixer Network for Person Search
PU-Transformer: Point Cloud Upsampling Transformer
QS-Craft: Learning to Quantize, Scrabble and Craft for Conditional Human Motion Animation
RaftMLP: How Much Can Be Done Without Attention and with Less Spatial Locality?
RA Loss: Relation-Aware Loss for Robust Person Re-identification
Rethinking Online Knowledge Distillation with Multi-exits
Revisiting Unsupervised Domain Adaptation Models: A Smoothness Perspective
RGB Road Scene Material Segmentation
Robust Human Matting via Semantic Guidance
Robustizing Object Detection Networks Using Augmented Feature Pooling
Scale Adaptive Fusion Network for RGB-D Salient Object Detection
SCOAD: Single-Frame Click Supervision for Online Action Detection
Self-distilled Vision Transformer for Domain Generalization
Self-supervised Augmented Patches Segmentation for Anomaly Detection
Self-supervised Learning with Multi-view Rendering for 3D Point Cloud Analysis

« prev Page 3/4 next »