论文/代码速递2022.10.19!

2022-12-11 12:57:06 浏览数 (1)

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

ECCV2022论文和代码整理:https://github.com/DWCTOD/ECCV2022-Papers-with-Code-Demo

最新成果demo展示:

标题:CLIFF: Carrying Location Information in Full Frames into Human Pose and Shape Estimation

代码:https://github.com/huawei-noah/noah-research/tree/master/CLIFF

论文:https://arxiv.org/abs/2208.00571

自顶向下的方法在3D人体姿势和形状估计领域占据主导地位,因为它们与人体检测分离,允许研究人员专注于核心问题。然而,裁剪是它们的第一步,从一开始就丢弃了位置信息,这使得它们无法在原始相机坐标系中准确预测全局旋转。为了解决这个问题,我们建议在这个任务中携带全帧位置信息(CLIFF)。具体来说,我们通过将裁剪的图像特征与其边界框信息连接起来,向CLIFF提供更全面的特征。我们在更宽的全帧视野下计算2D重投影损失,采用与在图像中投影的人相似的投影过程。CLIFF由全球位置感知信息提供并监督,它直接预测全球旋转以及更精确的关节姿势。此外,我们提出了一种基于CLIFF的伪地面真值注释器,它为野外二维数据集提供高质量的三维注释,并为基于回归的方法提供关键的全面监督。对流行基准测试的大量实验表明,CLIFF的表现明显优于现有技术,并在AGORA排行榜上排名第一(SMPL算法跟踪)。

最新论文整理

ECCV2022

Updated on : 19 Oct 2022
total number : 5

ARAH: Animatable Volume Rendering of Articulated Human SDFs

  • 论文/Paper: http://arxiv.org/pdf/2210.10036
  • 代码/Code: None

Towards Efficient and Effective Self-Supervised Learning of Visual Representations

  • 论文/Paper: http://arxiv.org/pdf/2210.09866
  • 代码/Code: https://github.com/val-iisc/effssl

On-the-go Reflectance Transformation Imaging with Ordinary Smartphones

  • 论文/Paper: http://arxiv.org/pdf/2210.09821
  • 代码/Code: None

Homogeneous Multi-modal Feature Fusion and Interaction for 3D Object Detection

  • 论文/Paper: http://arxiv.org/pdf/2210.09615
  • 代码/Code: None

Scaling Adversarial Training to Large Perturbation Bounds

  • 论文/Paper: http://arxiv.org/pdf/2210.09852
  • 代码/Code: None

CVPR2022

NeurIPS

Updated on : 19 Oct 2022
total number : 4

How Would The Viewer Feel? Estimating Wellbeing From Video Scenarios

  • 论文/Paper: http://arxiv.org/pdf/2210.10039
  • 代码/Code: https://github.com/hendrycks/emodiversity

Decoupling Features in Hierarchical Propagation for Video Object Segmentation

  • 论文/Paper: http://arxiv.org/pdf/2210.09782
  • 代码/Code: https://github.com/z-x-yang/AOT.

HUMANISE: Language-conditioned Human Motion Generation in 3D Scenes

  • 论文/Paper: http://arxiv.org/pdf/2210.09729
  • 代码/Code: None

Hierarchical Normalization for Robust Monocular Depth Estimation

  • 论文/Paper: http://arxiv.org/pdf/2210.09670
  • 代码/Code: None

0 人点赞