ECCV 2022

1,670 papers

Top Research Directions

Research directions most represented by ECCV 2022 papers.

Generalizing from Synthetic Data to the Real World 30
Exploiting Temporal Redundancy in Video Neural Networks 23
Addressing Inductive Bias Deficiency in Vision Transformers 21
Cross Task Generalization in Image Restoration 18
Scaling Visual Attention Beyond Quadratic Complexity 18
Preserving Tiny Features Against Spatial Resolution Degradation 16
Decoupled Spatiotemporal Attention in Transformer Architectures 15
Cascaded Error Propagation in 2D to 3D Lifting 14
Mitigating Error Propagation in Pseudo Labeling 14
Bypassing the Computational Bottleneck of Adversarial Training 14
Disentangling Attribute Control in Generative Models 13
Overcoming Local Receptive Field Constraints in Visual Models 13
Deep Learning on Unstructured 3D Point Clouds 13
Clean Accuracy Degradation in Machine Learning Defenses 13
Dense Prediction Adaptation 12

Top Authors

AuthorPapers
Giovanni Maria Farinella26
Dacheng Tao15
Luc Van Gool14
Ziwei Liu13
Chen Change Loy12
Alan Yuille10
Jing Zhang10
Rongrong Ji10
Shouhong Ding10
Yu Qiao10
Hongsheng Li9
Xiang Bai9
Boxin Shi8
Nicu Sebe8
Shuicheng YAN8
Wanli Ouyang8
Xiatian Zhu8
Zhangyang Wang8
Zhe Lin8
Errui Ding7

Impactful Papers

Papers from ECCV 2022 that shaped research clusters.

Foundational

Influential

Papers

Title
Power Efficient Video Super-Resolution on Mobile NPUs with Deep Learning, Mobile AI & AIM 2022 Challenge: Report
PPT: Token-Pruned Pose Transformer for Monocular and Multi-view Human Pose Estimation
Practical and Scalable Desktop-Based High-Quality Facial Capture
Predicting Is Not Understanding: Recognizing and Addressing Underspecification in Machine Learning
Prediction-Guided Distillation for Dense Object Detection
PREF: Predictability Regularized Neural Motion Fields
PressureVision: Estimating Hand Pressure from a Single RGB Image
Pre-training Strategies and Datasets for Facial Representation Learning
PreTraM: Self-supervised Pre-training via Connecting Trajectory and Map
PRIF: Primary Ray-Based Implicit Function
PRIME: A Few Primitives Can Boost Robustness to Common Corruptions
Primitive-Based Shape Abstraction via Nonparametric Bayesian Inference
Prior-Guided Adversarial Initialization for Fast Adversarial Training
Prior Knowledge Guided Unsupervised Domain Adaptation
Privacy-Preserving Action Recognition via Motion Difference Quantization
Privacy-Preserving Face Recognition with Learnable Privacy Budgets in Frequency Domain
PrivHAR: Recognizing Human Actions from Privacy-Preserving Lens
Projective Parallel Single-Pixel Imaging to Overcome Global Illumination in 3D Structure Light Scanning
PromptDet: Towards Open-Vocabulary Detection Using Uncurated Images
Prompting Visual-Language Models for Efficient Video Understanding
ProposalContrast: Unsupervised Pre-training for LiDAR-Based 3D Object Detection
Proposal-Free Temporal Action Detection via Global Segmentation Mask Learning
Prototype-Guided Continual Adaptation for Class-Incremental Unsupervised Domain Adaptation
Prototypical Contrast Adaptation for Domain Adaptive Semantic Segmentation
Prune Your Model Before Distill It
PseCo: Pseudo Labeling and Consistency Training for Semi-Supervised Object Detection
PseudoAugment: Learning to Use Unlabeled Data for Data Augmentation in Point Clouds
PseudoClick: Interactive Image Segmentation with Click Imitation
PS-NeRF: Neural Inverse Rendering for Multi-view Photometric Stereo
PSS: Progressive Sample Selection for Open-World Visual Representation Learning
P-STMO: Pre-trained Spatial Temporal Many-to-One Model for 3D Human Pose Estimation
PT4AL: Using Self-supervised Pretext Tasks for Active Learning
PTQ4ViT: Post-training Quantization for Vision Transformers with Twin Uniform Quantization
PTSEFormer: Progressive Temporal-Spatial Enhanced TransFormer Towards Video Object Detection
Pure Transformer with Integrated Experts for Scene Text Recognition
Q-FW: A Hybrid Classical-Quantum Frank-Wolfe for Quadratic Binary Optimization
QISTA-ImageNet: A Deep Compressive Image Sensing Framework Solving lq-Norm Optimization Problem
QISTA-ImageNet: A Deep Compressive Image Sensing Framework Solving ℓ q-Norm Optimization Problem
Quantized GAN for Complex Music Generation from Dance Videos
Quantum Motion Segmentation
Quasi-Balanced Self-Training on Noise-Aware Synthesis of Object Point Clouds for Closing Domain Gap
R2L: Distilling Neural Radiance Field to Neural Light Field for Efficient Novel View Synthesis
Radatron: Accurate Detection Using Multi-resolution Cascaded MIMO Radar
RA-Depth: Resolution Adaptive Self-supervised Monocular Depth Estimation
RadioTransformer: A Cascaded Global-Focal Transformer for Visual Attention-Guided Disease Classification
RamGAN: Region Attentive Morphing GAN for Region-Level Makeup Transfer
RankSeg: Adaptive Pixel Classification with Image Category Ranking for Segmentation
RAWtoBit: A Fully End-to-end Camera ISP Network
Rayleigh EigenDirections (REDs): Nonlinear GAN Latent Space Traversals for Multidimensional Features
RayTran: 3D Pose Estimation and Shape Reconstruction of Multiple Objects from Videos with Ray-Traced Transformers

« prev Page 24/34 next »