CVPR2022论文速递（2022.5.13）！共11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 : 13 May 2022

total number : 11

目标检测 / Object Detection - 1 篇

Group R-CNN for Weakly Semi-supervised Object Detection with Points

标题：用于弱半监督目标检测的Group R-CNN

• 论文/Paper: http://arxiv.org/pdf/2205.05920
• 代码/Code: https://github.com/jshilong/grouprcnn

语义分割/Segmentation - 1 篇

Delving into High-Quality Synthetic Face Occlusion Segmentation Datasets

标题：深入研究高质量的合成面部遮挡细分数据集

• 论文/Paper: http://arxiv.org/pdf/2205.06218
• 代码/Code: https://github.com/kennyvoo/face-occlusion-generation

GAN/ - 1 篇

Tensor-based Emotion Editing in the StyleGAN Latent Space

标题：Stylegan潜在空间中基于张量的情感编辑

• 论文/Paper: http://arxiv.org/pdf/2205.06102
• 代码/Code: None

超分/Super-Resolution - 1 篇

Blueprint Separable Residual Network for Efficient Image Super-Resolution

标题：蓝图可分离剩余网络，用于有效图像超分辨率

• 论文/Paper: http://arxiv.org/pdf/2205.05996
• 代码/Code: https://github.com/xiaom233/BSRN.

人脸相关 / Face - 1 篇

Delving into High-Quality Synthetic Face Occlusion Segmentation Datasets

标题：深入研究高质量的合成面部遮挡细分数据集

• 论文/Paper: http://arxiv.org/pdf/2205.06218
• 代码/Code: https://github.com/kennyvoo/face-occlusion-generation

点云/Point Clouds - 1 篇

Surface Representation for Point Clouds

标题：点云的表面表示

• 论文/Paper: http://arxiv.org/pdf/2205.05740
• 代码/Code: https://github.com/hancyran/RepSurf

其他/Other - 6 篇

Topologically-Aware Deformation Fields for Single-View 3D Reconstruction

标题：单视图3D重建的拓扑意识的变形场

• 论文/Paper: http://arxiv.org/pdf/2205.06267
• 代码/Code: None

3D Moments from Near-Duplicate Photos

标题：近乎近图片的3D时刻

• 论文/Paper: http://arxiv.org/pdf/2205.06255
• 代码/Code: None

What's in a Caption? Dataset-Specific Linguistic Diversity and Its Effect on Visual Description Models and Metrics

标题：标题中有什么？数据集特定的语言多样性及其对视觉描述模型和指标的影响

• 论文/Paper: http://arxiv.org/pdf/2205.06253
• 代码/Code: None

Infrared Invisible Clothing:Hiding from Infrared Detectors at Multiple Angles in Real World

标题：红外隐形服装：在现实世界中以多个角度躲藏在红外探测器上

• 论文/Paper: http://arxiv.org/pdf/2205.05909
• 代码/Code: None

Weakly-Supervised Action Detection Guided by Audio Narration

标题：由音频叙述指导的弱监督的动作检测

• 论文/Paper: http://arxiv.org/pdf/2205.05895
• 代码/Code: None

Diverse Video Generation from a Single Video

标题：单个视频的多样化视频生成

• 论文/Paper: http://arxiv.org/pdf/2205.05725
• 代码/Code: None