CVPR2022論文速遞（2022.5.17）！共6篇！
Updated on : 17 May 2022
total number : 6
語義分割/Segmentation - 1 篇
Deep Spectral Methods: A Surprisingly Strong Baseline for Unsupervised Semantic Segmentation and Localization
標(biāo)題：深度頻譜方法：無監(jiān)督語義分割和定位的令人驚訝的強(qiáng)基線
論文/Paper: http://arxiv.org/pdf/2205.07839
代碼/Code: None
Transformers - - 1 篇
Video Frame Interpolation with Transformer
標(biāo)題：基于Transformer的視頻框架插值
論文/Paper: http://arxiv.org/pdf/2205.07230
代碼/Code: None
視頻插幀/Frame Interpolation - 1 篇
Video Frame Interpolation with Transformer
標(biāo)題：基于Transformer的視頻框架插值
論文/Paper: http://arxiv.org/pdf/2205.07230
代碼/Code: None
其他/Other - 4 篇
FvOR: Robust Joint Shape and Pose Optimization for Few-view Object Reconstruction
標(biāo)題：FVOR：穩(wěn)健的關(guān)節(jié)形狀和姿勢優(yōu)化，用于重建對象
論文/Paper: http://arxiv.org/pdf/2205.07763
代碼/Code: https://github.com/zhenpeiyang/FvOR/
GLaMa: Joint Spatial and Frequency Loss for General Image Inpainting
標(biāo)題：GLaMa：用于通用的圖像修復(fù)的聯(lián)合空間和頻率損失
論文/Paper: http://arxiv.org/pdf/2205.07162
代碼/Code: None
SaiNet: Stereo aware inpainting behind objects with generative networks
標(biāo)題：SaiNet：立體聲意識到具有生成網(wǎng)絡(luò)的對象后面的涂料
論文/Paper: http://arxiv.org/pdf/2205.07014
代碼/Code: None
RiCS: A 2D Self-Occlusion Map for Harmonizing Volumetric Objects
標(biāo)題：RiCS：用于協(xié)調(diào)體積對象的2D自遮擋圖
論文/Paper: http://arxiv.org/pdf/2205.06975
代碼/Code: None