整理：AI算法与图像处理

CVPR2022论文和代码整理：https://github.com/DWCTOD/CVPR2022-Papers-with-Code-Demo

欢迎关注：

Transforming Model Prediction for Tracking

• 论文/Paper: http://arxiv.org/pdf/2203.11192

• 代码/Code: None

DiffPoseNet: Direct Differentiable Camera Pose Estimation

• 论文/Paper: http://arxiv.org/pdf/2203.11174

• 代码/Code: None

Not All Points Are Equal: Learning Highly Efficient Point-based Detectors for 3D LiDAR Point Clouds

• 论文/Paper: http://arxiv.org/pdf/2203.11139

• 代码/Code: None

No Pain, Big Gain: Classify Dynamic Point Cloud Sequences with Static Models by Fitting Feature-level Space-time Surfaces

• 论文/Paper: http://arxiv.org/pdf/2203.11113

• 代码/Code: None

MixFormer: End-to-End Tracking with Iterative Mixed Attention

• 论文/Paper: http://arxiv.org/pdf/2203.11082

• 代码/Code: None

MonoDTR: Monocular 3D Object Detection with Depth-Aware Transformer

• 论文/Paper: http://arxiv.org/pdf/2203.10981

• 代码/Code: None

Unified Multivariate Gaussian Mixture for Efficient Neural Image Compression

• 论文/Paper: http://arxiv.org/pdf/2203.10897

• 代码/Code: None

Revisiting Domain Generalized Stereo Matching Networks from a Feature Consistency Perspective

• 论文/Paper: http://arxiv.org/pdf/2203.10887

• 代码/Code: None

ELIC: Efficient Learned Image Compression with Unevenly Grouped Space-Channel Contextual Adaptive Coding

• 论文/Paper: http://arxiv.org/pdf/2203.10886

• 代码/Code: None

RGB-Depth Fusion GAN for Indoor Depth Completion

• 论文/Paper: http://arxiv.org/pdf/2203.10856

• 代码/Code: None

Hyperbolic Vision Transformers: Combining Improvements in Metric Learning

• 论文/Paper: http://arxiv.org/pdf/2203.10833

• 代码/Code: None

ViM: Out-Of-Distribution with Virtual-logit Matching

• 论文/Paper: http://arxiv.org/pdf/2203.10807

• 代码/Code: None

Delving into the Estimation Shift of Batch Normalization in a Network

• 论文/Paper: http://arxiv.org/pdf/2203.10778

• 代码/Code: None

Tree Energy Loss: Towards Sparsely Annotated Semantic Segmentation

• 论文/Paper: http://arxiv.org/pdf/2203.10739

• 代码/Code: None

HP-Capsule: Unsupervised Face Part Discovery by Hierarchical Parsing Capsule Network

• 论文/Paper: http://arxiv.org/pdf/2203.10699

• 代码/Code: None

Unsupervised Domain Adaptation for Nighttime Aerial Tracking

• 论文/Paper: http://arxiv.org/pdf/2203.10541

• 代码/Code: None

Depth Estimation by Combining Binocular Stereo and Monocular Structured-Light

• 论文/Paper: http://arxiv.org/pdf/2203.10493

• 代码/Code: None

SimAN: Exploring Self-Supervised Representation Learning of Scene Text via Similarity-Aware Normalization

• 论文/Paper: http://arxiv.org/pdf/2203.10492

• 代码/Code: None

TVConv: Efficient Translation Variant Convolution for Layout-aware Visual Processing

• 论文/Paper: http://arxiv.org/pdf/2203.10489

• 代码/Code: None

Portrait Eyeglasses and Shadow Removal by Leveraging 3D Synthetic Data

• 论文/Paper: http://arxiv.org/pdf/2203.10474

• 代码/Code: None

CLRNet: Cross Layer Refinement Network for Lane Detection

• 论文/Paper: http://arxiv.org/pdf/2203.10350

• 代码/Code: None

Voxel Set Transformer: A Set-to-Set Approach to 3D Object Detection from Point Clouds

• 论文/Paper: http://arxiv.org/pdf/2203.10314

• 代码/Code: None

DirecFormer: A Directed Attention in Transformer Approach to Robust Action Recognition

• 论文/Paper: http://arxiv.org/pdf/2203.10233

• 代码/Code: None

SwinTextSpotter: Scene Text Spotting via Better Synergy between Text Detection and Text Recognition

• 论文/Paper: http://arxiv.org/pdf/2203.10209

• 代码/Code: None

Discovering Objects that Can Move

• 论文/Paper: http://arxiv.org/pdf/2203.10159

• 代码/Code: None