CVPR2022論文速遞（2022.5.25）！共8篇！

Updated on : ?25 May 2022

total number : 8

目標跟蹤/Object Tracking - 1 篇

Ranking-Based Siamese Visual Tracking

標題：基于排名的Siamese視覺跟蹤

• 論文/Paper: http://arxiv.org/pdf/2205.11761

• 代碼/Code: https://github.com/sansanfree/rbo

姿態(tài)估計/Pose Estimation - 1 篇

OnePose: One-Shot Object Pose Estimation without CAD Models

標題：OnePose：無CAD模型的one-shot對象姿勢估計

• 論文/Paper: http://arxiv.org/pdf/2205.12257

• 代碼/Code: None

NeRF - 1 篇

StylizedNeRF: Consistent 3D Scene Stylization as Stylized NeRF via 2D-3D Mutual Learning

標題：Stylizednerf：一致的3D場景風格作為通過2D-3D相互學習的風格化NERF

• 論文/Paper: http://arxiv.org/pdf/2205.12183

• 代碼/Code: None

其他/Other - 5 篇

Differentiable Dynamics for Articulated 3d Human Motion Reconstruction

標題：鉸接3D人類運動重建的可區(qū)分動力

• 論文/Paper: http://arxiv.org/pdf/2205.12256

• 代碼/Code: None

Context Attention Network for Skeleton Extraction

標題：骨架提取的上下文注意網(wǎng)絡(luò)

• 論文/Paper: http://arxiv.org/pdf/2205.12066

• 代碼/Code: None

SCVRL: Shuffled Contrastive Video Representation Learning

標題：SCVRL：Shuffled的對比視頻表示學習

• 論文/Paper: http://arxiv.org/pdf/2205.11710

• 代碼/Code: None

TransforMatcher: Match-to-Match Attention for Semantic Correspondence

標題：轉(zhuǎn)換器：語義通信的匹配點注意

• 論文/Paper: http://arxiv.org/pdf/2205.11634

• 代碼/Code: None

Exploring Concept Contribution Spatially: Hidden Layer Interpretation with Spatial Activation Concept Vector

標題：在空間上探索概念的貢獻：隱藏的圖層解釋與空間激活概念向量

• 論文/Paper: http://arxiv.org/pdf/2205.11511

• 代碼/Code: https://github.com/antonotnawang/spatial-activation-concept-vector

•