极市一直非常关注CVPR，在每年都会进行CVPR相关的论文资源整理，包括论文解读、代码、技术直播、分方向盘点、最佳论文汇总等，也得到了许多开发者的支持。在Github上，极市的CVPR盘点已经收获了5.7K star。在今年，我们也会对CVPR2021进行实时跟进。
Github项目地址：
https://github.com/extreme-assistant/CVPR2021-Paper-Code-Interpretation
以下是我们对本次CVPR2021最新论文的整理：
CVPR2021最新信息及论文下载（Papers/Codes/Project/PaperReading／Demos/直播分享／论文分享会等）
官网链接：http://cvpr2021.thecvf.com
时间：2021年6月19日-6月25日
论文接收公布时间：2021年2月28日
相关问题：
目录
1. CVPR2021接受论文/代码（持续更新）
2. To do list
1.CVPR2021接受论文/代码(持续更新)
61.Coarse-Fine Networks for Temporal Activity Detection in Videos(用于视频中的时间活动检测的粗细网络)
60.Instance Localization for Self-supervised Detection Pretraining（自监督检测预训练的实例定位）
paper|code
59.Weakly-supervised Grounded Visual Question Answering using Capsules（使用胶囊进行弱监督的地面视觉问答）
58.FLAVR: Flow-Agnostic Video Representations for Fast Frame Interpolation（FLAVR：用于快速帧插值的与流无关的视频表示）
paper|code|project
57.4D Panoptic LiDAR Segmentation（4D全景LiDAR分割）
paper
56.Dogfight: Detecting Drones from Drone Videos（从无人机视频中检测无人机）
55.Multiple Instance Active Learning for Object Detection（用于对象检测的多实例主动学习）
paper|code
54.Re-labeling ImageNet: from Single to Multi-Labels, from Global to Localized Labels（重新标记ImageNet：从单标签到多标签，从全局标签到本地标签）
paper|code
53.Probabilistic Embeddings for Cross-Modal Retrieval（跨模态检索的概率嵌入）
paper
52.PLOP: Learning without Forgetting for Continual Semantic Segmentation（PLOP：学习而不会忘记连续的语义分割）
paper
51.Rainbow Memory: Continual Learning with a Memory of Diverse Samples（不断学习与多样本的记忆）
50.Exploiting Spatial Dimensions of Latent in GAN for Real-time Image Editing（利用GAN中潜在的空间维度进行实时图像编辑）
48.Rethinking Channel Dimensions for Efficient Model Design(重新考虑通道尺寸以进行有效的模型设计)
paper|code
46.Reconsidering Representation Alignment for Multi-view Clustering(重新考虑多视图聚类的表示对齐方式)
45.Self-supervised Simultaneous Multi-Step Prediction of Road Dynamics and Cost Map(道路动力学和成本图的自监督式多步同时预测)
44.IIRC: Incremental Implicitly-Refined Classification(增量式隐式定义的分类)
paper|project
43.Fair Attribute Classification through Latent Space De-biasing(通过潜在空间去偏的公平属性分类)
paper|code|project
42.Information-Theoretic Segmentation by Inpainting Error Maximization(修复误差最大化的信息理论分割)
paper
41.Probabilistic Tracklet Scoring and Inpainting for Multiple Object Tracking(多目标跟踪的概率小波计分和修复)
paper
40.Few-shot Open-set Recognition by Transformation Consistency(转换一致性很少的开放集识别)
39.Meta Batch-Instance Normalization for Generalizable Person Re-Identification(通用批处理人员重新标识的元批实例规范化)
paper
37.UC2: Universal Cross-lingual Cross-modal Vision-and-Language Pretraining(UC2：通用跨语言跨模态视觉和语言预培训)
36.Less is More: CLIPBERT for Video-and-Language Learning via Sparse Sampling(T通过稀疏采样进行视频和语言学习)
paper|code
35.3D Graph Anatomy Geometry-Integrated Network for Pancreatic Mass Segmentation, Diagnosis, and Quantitative Patient Management(用于胰腺肿块分割，诊断和定量患者管理的3D图形解剖学几何集成网络)
34.Deep Lesion Tracker: Monitoring Lesions in 4D Longitudinal Imaging Studies(深部病变追踪器：在4D纵向成像研究中监控病变)
paper
33.Automatic Vertebra Localization and Identification in CT by Spine Rectification and Anatomically-constrained Optimization(通过脊柱矫正和解剖学约束优化在CT中自动进行椎骨定位和识别)
paper
31.3D CNNs with Adaptive Temporal Feature Resolutions(具有自适应时间特征分辨率的3D CNN)
paper
30.KeepAugment: A Simple Information-Preserving Data Augmentation(一种简单的保存信息的数据扩充)
paper
29.AttentiveNAS: Improving Neural Architecture Search via Attentive(通过注意力改善神经架构搜索)
paper
28.Hijack-GAN: Unintended-Use of Pretrained, Black-Box GANs(Hijack-GAN：意外使用经过预训练的黑匣子GAN)
paper
27.D-NeRF: Neural Radiance Fields for Dynamic Scenes(D-NeRF：动态场景的神经辐射场)
project
26.Encoding in Style: a StyleGAN Encoder for Image-to-Image Translation(样式编码：用于图像到图像翻译的StyleGAN编码器)
project
25.A 3D GAN for Improved Large-pose Facial Recognition(用于改善大姿势面部识别的3D GAN)
paper
24.Rotation Equivariant Siamese Networks for Tracking（旋转等距连体网络进行跟踪）
paper
23.Exploring Complementary Strengths of Invariant and Equivariant Representations for Few-Shot Learning(探索少量学习的不变表示形式和等变表示形式的互补强度)
22.Open-world object detection(开放世界中的目标检测)
code
21.Multi-Stage Progressive Image Restoration(多阶段渐进式图像复原)
paper|code
20.Weakly Supervised Learning of Rigid 3D Scene Flow(刚性3D场景流的弱监督学习)
paper|code|project
19.PREDATOR: Registration of 3D Point Clouds with Low Overlap(预测器：低重叠的3D点云的注册)
paper|code|project
18.Sequential Graph Convolutional Network for Active Learning(主动学习的顺序图卷积网络)
paper
17.Multiresolution Knowledge Distillation for Anomaly Detection(用于异常检测的多分辨率知识蒸馏)
paper
16.Positive-Unlabeled Data Purification in the Wild for Object Detection(野外检测对象的阳性无标签数据提纯)
15.Data-Free Knowledge Distillation For Image Super-Resolution(DAFL算法的SR版本)
14.Manifold Regularized Dynamic Network Pruning（动态剪枝的过程中考虑样本复杂度与网络复杂度的约束）
13.Distilling Object Detectors via Decoupled Features（前景背景分离的蒸馏技术）
12.Inverting the Inherence of Convolution for Visual Recognition（颠倒卷积的固有性以进行视觉识别）
11.Representative Batch Normalization with Feature Calibration（具有特征校准功能的代表性批量归一化）
10.PointFlow: Flowing Semantics Through Points for Aerial Image Segmentation(语义流经点以进行航空图像分割)
9.Learning the Superpixel in a Non-iterative and Lifelong Manner(以非迭代和终身的方式学习超像素)
8.RepVGG: Making VGG-style ConvNets Great Again
paper|code
解读：RepVGG：极简架构，SOTA性能，让VGG式模型再次伟大
7.Transformer Interpretability Beyond Attention Visualization(注意力可视化之外的Transformer可解释性)
paper|code
6.UP-DETR: Unsupervised Pre-training for Object Detection with Transformers
paper
解读：无监督预训练检测器
5.Pre-Trained Image Processing Transformer(底层视觉预训练模型)
paper
4.ReNAS: Relativistic Evaluation of Neural Architecture Search(NAS predictor当中ranking loss的重要性)
paper
3.AdderSR: Towards Energy Efficient Image Super-Resolution(将加法网路应用到图像超分辨率中)
paper|code
解读：华为开源加法神经网络
2.Learning Student Networks in the Wild（一种不需要原始训练数据的模型压缩和加速技术）
paper|code
解读：华为诺亚方舟实验室提出无需数据网络压缩技术
1.HourNAS: Extremely Fast Neural Architecture Search Through an Hourglass Lens（降低NAS的成本）
paper
2.To do list
- CVPR2021论文解读
- CVPR2021 Oral
- CVPR2021论文分享