ECCV2022 &CVPR2022论文速递2022.7.27！

2022-12-11 11:23:47 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：

ECCV2022 | ChunkyGAN 又一个狠活

论文：https://dcgi.fel.cvut.cz/home/sykorad/Subrtova22-ECCV.pdf 代码：https://github.com/futscdav/Chunkmogrify

ECCV2022 汇总：https://github.com/DWCTOD/ECCV2022-Papers-with-Code-Demo/

摘要：

我们提出了 ChunkyGAN——一种使用生成对抗网络对图像进行建模和编辑的新范式。与先前寻求输入图像的全局潜在表示的技术不同，我们的方法将输入图像细分为一组较小的组件（块），这些组件可以手动或使用预训练的分割网络自动指定。对于每个块，由于约束数量较少，生成网络的潜在代码在本地以更高的准确度进行估计。此外，在潜码优化过程中，可以进一步细化分割以提高匹配质量。这个过程可以对原始图像进行高质量的投影，并实现以前的方法难以实现的空间解缠结。为了证明我们方法的优势，我们在各种图像编辑场景中对其进行了定量和定性评估，这些场景受益于该方法的更高重建质量和局部性质。我们的方法足够灵活，甚至可以操作使用全局技术难以重建的域外图像

最新论文整理

ECCV2022

Updated on : 27 Jul 2022
total number : 14

NewsStories: Illustrating articles with visual summaries

  • 论文/Paper: http://arxiv.org/pdf/2207.13061
  • 代码/Code: https://github.com/newsstoriesdata/newsstories.github.io

Monocular 3D Object Detection with Depth from Motion

  • 论文/Paper: http://arxiv.org/pdf/2207.12988
  • 代码/Code: https://github.com/tai-wang/depth-from-motion

Efficient One Pass Self-distillation with Zipf's Label Smoothing

  • 论文/Paper: http://arxiv.org/pdf/2207.12980
  • 代码/Code: https://github.com/megvii-research/zipfls

Tracking Every Thing in the Wild

  • 论文/Paper: http://arxiv.org/pdf/2207.12978
  • 代码/Code: None

Contextual Text Block Detection towards Scene Text Understanding

  • 论文/Paper: http://arxiv.org/pdf/2207.12955
  • 代码/Code: None

AlignSDF: Pose-Aligned Signed Distance Fields for Hand-Object Reconstruction

  • 论文/Paper: http://arxiv.org/pdf/2207.12909
  • 代码/Code: None

Compositional Human-Scene Interaction Synthesis with Semantic Control

  • 论文/Paper: http://arxiv.org/pdf/2207.12824
  • 代码/Code: https://github.com/zkf1997/coins

Static and Dynamic Concepts for Self-supervised Video Representation Learning

  • 论文/Paper: http://arxiv.org/pdf/2207.12795
  • 代码/Code: None

Learning Visual Representation from Modality-Shared Contrastive Language-Image Pre-training

  • 论文/Paper: http://arxiv.org/pdf/2207.12661
  • 代码/Code: https://github.com/hxyou/msclip

Learning Hierarchy Aware Features for Reducing Mistake Severity

  • 论文/Paper: http://arxiv.org/pdf/2207.12646
  • 代码/Code: https://github.com/07agarg/haf

Translating a Visual LEGO Manual to a Machine-Executable Plan

  • 论文/Paper: http://arxiv.org/pdf/2207.12572
  • 代码/Code: None

3D Shape Sequence of Human Comparison and Classification using Current and Varifolds

  • 论文/Paper: http://arxiv.org/pdf/2207.12485
  • 代码/Code: https://github.com/cristal-3dsam/humancomparisonvarifolds

Semi-Leak: Membership Inference Attacks Against Semi-supervised Learning

  • 论文/Paper: http://arxiv.org/pdf/2207.12535
  • 代码/Code: https://github.com/xinleihe/semi-leak

Trainability Preserving Neural Structured Pruning

  • 论文/Paper: http://arxiv.org/pdf/2207.12534
  • 代码/Code: https://github.com/mingsun-tse/tpp

CVPR2022

Updated on : 27 Jul 2022
total number : 1

V^2L: Leveraging Vision and Vision-language Models into Large-scale Product Retrieval

  • 论文/Paper: http://arxiv.org/pdf/2207.12994
  • 代码/Code: https://github.com/WangWenhao0716/V2L
编程算法 http https github 网络安全

0 人点赞