Stare-at-What-You-See阅读笔记 Stare at What You See: Masked Image Modeling without Reconstruction 阅读笔记 论文标题 论文地址:2211.08887.pdf (arxiv.org) 该论文收录于CVPR2023 亮点 在Masked Image Modeling(MIM)中提出MaskAlign使得不用在masked 2024-03-05 #masked image modeling 自监督学习
SVFormer: Semi-supervised Video Transformer for Action Recognition SVFormer: Semi-supervised Video Transformer for Action Recognition 阅读笔记 论文地址:SVFormer: Semi-supervised Video Transformer for Action Recognition 亮点: 提出一种新颖的数据增强策略——Tube TokenMix(TTMix)。通过利用 2024-03-03 论文粗读 #动作识别
Robust-Rcovery-of-Subspace-Structure-by-Low-Rank-Representation精读笔记 Robust-Rcovery-of-Subspace-Structure-by-Low-Rank-Representation精读笔记 背景 论文链接:Robust-Rcovery-of-Subspace-Structure-by-Low-Rank-Representation RPCA(鲁棒主成分分析)把一个观测到的数据分解成一个干净的数据矩阵和一个稀疏的噪声矩阵的和的形式。 \[ X 2023-06-11 论文精读 #LRR低秩表示
Image Super-Resolution Using Very Deep Residual Channel Attention Networks 精读笔记 Image Super-Resolution Using Very Deep Residual Channel Attention Networks 精读笔记 背景 这是一篇收录在ECCV2018的一篇单幅图像超分辨率论文 论文链接:Image Super-Resolution Using Very Deep Residual Channel Attention Networks 2023-06-09 论文精读 #RCAN单幅图像超分辨率