MrPhu / CVPR2021_CVPapers

Phong-Phu Le - National Cheng-Kung University

Mr Phu Git-Pages

1. Paper Note: https://mrphu.github.io/Personal-Paper-Notes/
2. CVPR2021 Collection: https://mrphu.github.io/CVPR2021_CVPapers/
3. WACV 2021 Collection: https://mrphu.github.io/WACV2021_CVPapers/

---

CVPR_2021 Computer Vision Paper Collection

1. Classification / Backbone enhancement

• ReXNet: Diminishing Representational Bottleneck on Convolutional Neural Network [Code]_CVPR2021
• Involution: Inverting the Inherence of Convolution for Visual Recognition [Code]_CVPR2021
• Coordinate Attention for Efficient Mobile Network Design [Code]_CVPR2021
• Inception Convolution with Efficient Dilation Search [Code]_CVPR2021
• RepVGG: Making VGG-style ConvNets Great Again [Code]_CVPR2021

2. Object Detection

2.1 Object Detection on COCO

• You Only Look One-level Feature_CVPR2021
• Multiple Instance Active Learning for Object Detection [Code]_CVPR2021
• Positive-Unlabeled Data Purification in the Wild for Object Detection_CVPR2021
• Depth from Camera Motion and Object Detection_CVPR2021
• Towards Open World Object Detection [Code]_CVPR2021
• General Instance Distillation for Object Detection_CVPR2021
• Distilling Object Detectors via Decoupled Features_CVPR2021
• MeGA-CDA: Memory Guided Attention for Category-Aware Unsupervised Domain Adaptive Object Detection_CVPR2021
• Informative and Consistent Correspondence Mining for Cross-Domain Weakly Supervised Object Detection_CVPR2021
• Sparse R-CNN: End-to-End Object Detection with Learnable Proposals_CVPR2021
• OPANAS: One-Shot Path Aggregation Network Architecture Search for Object Detection [Code]_CVPR2021

2.2 Small Object Detection

• Semantic Relation Reasoning for Shot-Stable Few-Shot Object Detection_CVPR2021
• Dense Relation Distillation with Context-aware Aggregation for Few-Shot Object Detection_CVPR2021
• FSCE: Few-Shot Object Detection via Contrastive Proposal Encoding_CVPR2021

2.3 Multi-target Detection

• There is More than Meets the Eye: Self-Supervised Multi-Object Detection and Tracking with Sound by Distilling Multimodal Knowledge_CVPR2021

2.4 3D Target Detection

• Categorical Depth Distribution Network for Monocular 3D Object Detection_CVPR2021
• 3DIoUMatch: Leveraging IoU Prediction for Semi-Supervised 3D Object Detection_CVPR2021
• ST3D: Self-training for Unsupervised Domain Adaptation on 3D ObjectDetection_CVPR2021
• Depth-conditioned Dynamic Message Propagation for Monocular 3D Object Detection_CVPR2021

2.5 Rotating target detection

• Dense Label Encoding for Boundary Discontinuity Free Rotation Detection_CVPR2021

2.6 Target setting

• Unveiling the Potential of Structure-Preserving for Weakly Supervised Object Localization_CVPR2021

2.7 Dense object detection

• Generalized Focal Loss V2: Learning Reliable Localization Quality Estimation for Dense Object Detection [Code]_CVPR2021

3. Segmentation

• Information-Theoretic Segmentation by Inpainting Error Maximization_CVPR2021
• Simultaneously Localize, Segment and Rank the Camouflaged Objects_CVPR2021
• Capturing Omni-Range Context for Omnidirectional Segmentation_CVPR2021

3.1 Panoptic Segmentation

• Fully Convolutional Networks for Panoptic Segmentation [Code]_CVPR2021
• Cross-View Regularization for Domain Adaptive Panoptic Segmentation_CVPR2021

3.2 Instance segmentation

• Zero-Shot Instance Segmentation_CVPR2021
• Panorama segmentation_CVPR2021
• 4D Panoptic LiDAR Segmentation_CVPR2021
• Cross-View Regularization for Domain Adaptive Panoptic Segmentation_CVPR2021
• Part-aware Panoptic Segmentation_CVPR2021
• Toward Joint Thing-and-Stuff Mining for Weakly Supervised Panoptic Segmentation_CVPR2021

3.3 Semantic segmentation

• PLOP: Learning without Forgetting for Continual Semantic Segmentation_CVPR2021
• Towards Semantic Segmentation of Urban-Scale 3D Point Clouds: A Dataset, Benchmarks and Challenges_CVPR2021
• Multi-Source Domain Adaptation with Collaborative Learning for Semantic Segmentatio_CVPR2021
• Semi-supervised Domain Adaptation based on Dual-level Domain Mixing for Semantic Segmentation_CVPR2021
• Differentiable Multi-Granularity Human Representation Learning for Instance-Aware Human Semantic Parsing_CVPR2021
• Learning Statistical Texture for Semantic Segmentation_CVPR2021
• MetaCorrection: Domain-aware Meta Loss Correction for Unsupervised Domain Adaptation in Semantic Segmentation_CVPR2021
• Continual Semantic Segmentation via Repulsion-Attraction of Sparse and Disentangled Latent Representations_CVPR2021
• Semantic Segmentation for Real Point Cloud Scenes via Bilateral Augmentation and Adaptive Fusion_CVPR2021
• Rethinking BiSeNet For Real-time Semantic Segmentation_CVPR2021
• BBAM: Bounding Box Attribution Map for Weakly Supervised Semantic and Instance Segmentation_CVPR2021
• Anti-Adversarially Manipulated Attributions for Weakly and Semi-Supervised Semantic Segmentation_CVPR2021

3.4 Scene understanding/scene analysis

• Exploring Data Efficient 3D Scene Understanding with Contrastive Scene Contexts_CVPR2021
• Exploiting Edge-Oriented Reasoning for 3D Point-based Scene Graph Analysis_CVPR2021
• Probabilistic Modeling of Semantic Ambiguity for Scene Graph Generation_CVPR2021
• Monte Carlo Scene Search for 3D Scene Understanding_CVPR2021

3.5 Cutout

• Real-Time High Resolution Background Matting_CVPR2021

3.6 Video action segmentation

• Global2Local: Efficient Structure Search for Video Action Segmentation_CVPR2021

3.7 Sequential action segmentation

• Temporal Action Segmentation from Timestamp Supervision_CVPR2021

3.8 Radar segmentation

• Cylindrical and Asymmetrical 3D Convolution Networks for LiDAR Segmentation_CVPR2021

3.9 Video object segmentation

• Modular Interactive Video Object Segmentation:Interaction-to-Mask, Propagation and Difference-Aware Fusion_CVPR2021

3.10 Seg in Medical

• DoDNet: Learning to segment multi-organ and tumors from multiple partially labeled datasets_CVPR2021
• FedDG: Federated Domain Generalization on Medical Image Segmentation via Episodic Learning in Continuous Frequency Space [Code]_CVPR2021
• Deep Learning for Chest X-ray Analysis: A Survey_CVPR2021

4. Transformer in Visual

• UP-DETR: Unsupervised Pre-training for Object Detection with Transformers [Code]_CVPR2021
• Transformer Interpretability Beyond Attention Visualization [Code]_CVPR2021
• Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers_CVPR2021
• VisTR: End-to-End Video Instance Segmentation with Transformers [Code]_CVPR2021
• End-to-End Human Object Interaction Detection with HOI Transformers [Code]_CVPR2021

5. Tracking

• Rotation Equivariant Siamese Networks for Tracking_CVPR2021
• Graph Attention Tracking_CVPR2021

5.1 Multi-target tracking

• Probabilistic Tracklet Scoring and Inpainting for Multiple Object Tracking_CVPR2021
• Track to Detect and Segment: An Online Multi-Object Tracker_CVPR2021
• Multiple Object Tracking with Correlation Learning_CVPR2021
• Learning a Proposal Classifier for Multiple Object Tracking_CVPR2021

6. Anomaly Detection

• Multiresolution Knowledge Distillation for Anomaly Detection_CVPR2021

7. Data augmentation

• KeepAugment: A Simple Information-Preserving Data Augmentation_CVPR2021

8. GAN

• Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editin_CVPR2021
• Encoding in Style: a StyleGAN Encoder for Image-to-Image Translation_CVPR2021
• Hijack-GAN: Unintended-Use of Pretrained, Black-Box GANs_CVPR2021
• Image-to-image Translation via Hierarchical Style Disentanglement [Code]_CVPR2021
• Efficient Conditional GAN Transfer with Knowledge Propagation across Classes_CVPR2021
• Anycost GANs for Interactive Image Synthesis and Editing [Code]_CVPR2021
• TediGAN: Text-Guided Diverse Image Generation and Manipulation_CVPR2021
• Generative Hierarchical Features from Synthesizing Images_CVPR2021
• Teachers Do More Than Teach: Compressing Image-to-Image Models_CVPR2021
• PISE: Person Image Synthesis and Editing with Decoupled GAN_CVPR2021
• LOHO: Latent Optimization of Hairstyles via Orthogonalization_CVPR2021
• Image-to-image Translation via Hierarchical Style Disentanglement_CVPR2021
• CoMoGAN: continuous model-guided image-to-image translation_CVPR2021
• HumanGAN: A Generative Model of Humans Images_CVPR2021
• HistoGAN: Controlling Colors of GAN-Generated and Real Images via Color Histograms_CVPR2021
• DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network_CVPR2021
• Training Generative Adversarial Networks in One Stage_CVPR2021
• Closed-Form Factorization of Latent Semantics in GANs [Code]_CVPR2021

Good Respitories

Paper Summary

• Sorting CVPR2021 Papers: https://github.com/52CV/CVPR-2021-Papers
• CVPR2021 Paper_with_code: https://github.com/amusi/CVPR2021-Papers-with-Code