CVPR2022论文速递（2022.4.28）！共12篇！

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

Updated on : 28 Apr 2022

total number : 12

语义分割/Segmentation - 2 篇

Self-Supervised Learning of Object Parts for Semantic Segmentation

标题：对语义细分的对象零件的自我监督学习

• 论文/Paper: http://arxiv.org/pdf/2204.13101
• 代码/Code: None

MM-TTA: Multi-Modal Test-Time Adaptation for 3D Semantic Segmentation

标题：MM-TTA：3D语义分割的多模式测试时间适应

• 论文/Paper: http://arxiv.org/pdf/2204.12667
• 代码/Code: None

Transformers - - 1 篇

DearKD: Data-Efficient Early Knowledge Distillation for Vision Transformers

标题：dearkd：视觉变压器的数据有效的早期知识蒸馏

• 论文/Paper: http://arxiv.org/pdf/2204.12997
• 代码/Code: None

姿态估计/Pose Estimation - 1 篇

Coupled Iterative Refinement for 6D Multi-Object Pose Estimation

标题：6D多对象姿势估计的迭代精致

• 论文/Paper: http://arxiv.org/pdf/2204.12516
• 代码/Code: https://github.com/princeton-vl/Coupled-Iterative-Refinement.

点云/Point Clouds - 1 篇

Density-preserving Deep Point Cloud Compression

标题：具有密度的深点云压缩

• 论文/Paper: http://arxiv.org/pdf/2204.12684
• 代码/Code: None

其他/Other - 7 篇

Few-Shot Head Swapping in the Wild

标题：在野外换人的头部交换

• 论文/Paper: http://arxiv.org/pdf/2204.13100
• 代码/Code: None

Attention Consistency on Visual Corruptions for Single-Source Domain Generalization

标题：单一源域概括的视觉损坏的注意力一致性

• 论文/Paper: http://arxiv.org/pdf/2204.13091
• 代码/Code: None

Collaborative Learning for Hand and Object Reconstruction with Attention-guided Graph Convolution

标题：通过注意力引导的图形卷积，手和对象重建的协作学习

• 论文/Paper: http://arxiv.org/pdf/2204.13062
• 代码/Code: None

A Scalable Combinatorial Solver for Elastic Geometrically Consistent 3D Shape Matching

标题：可伸缩的组合求解器，用于弹性几何一致的3D形状匹配

• 论文/Paper: http://arxiv.org/pdf/2204.12805
• 代码/Code: http://github.com/paul0noah/sm-comb

Leveraging Unlabeled Data for Sketch-based Understanding

标题：利用未标记的数据进行基于草图的理解

• 论文/Paper: http://arxiv.org/pdf/2204.12522
• 代码/Code: None

Towards assessing agricultural land suitability with causal machine learning

标题：通过因果机学习评估农业土地适用性

• 论文/Paper: http://arxiv.org/pdf/2204.12956
• 代码/Code: None

Conformer and Blind Noisy Students for Improved Image Quality Assessment

标题：构象异构和盲目嘈杂的学生，以改进图像质量评估

• 论文/Paper: http://arxiv.org/pdf/2204.12819
• 代码/Code: None