CVPR2022論文速遞（2022.4.15）！共16篇！內(nèi)含2篇Oral！
Updated on : 15 Apr 2022
total number : 16
CVPR2022 Oral - 2 篇
Joint Forecasting of Panoptic Segmentations with Difference Attention ?(Oral)
標(biāo)題：差異關(guān)注聯(lián)合預(yù)測全景分割
論文/Paper: http://arxiv.org/pdf/2204.07157
代碼/Code: None
Deformable Sprites for Unsupervised Video Decomposition ?(Oral)
標(biāo)題：無監(jiān)督視頻分解的可變形Sprites
論文/Paper: http://arxiv.org/pdf/2204.07151
代碼/Code: None
目標(biāo)跟蹤/Object Tracking - 2 篇
BEHAVE: Dataset and Method for Tracking Human Object Interactions
標(biāo)題：BEHAVE：數(shù)據(jù)集和用于跟蹤人類對(duì)象交互的方法
論文/Paper: http://arxiv.org/pdf/2204.06950
代碼/Code: None
SoccerNet-Tracking: Multiple Object Tracking Dataset and Benchmark in Soccer Videos
標(biāo)題：SoccerNet跟蹤：多個(gè)對(duì)象跟蹤數(shù)據(jù)集和足球視頻中的基準(zhǔn)
論文/Paper: http://arxiv.org/pdf/2204.06918
代碼/Code: None
語義分割/Segmentation - 2 篇
Joint Forecasting of Panoptic Segmentations with Difference Attention ?(Oral)
標(biāo)題：差異關(guān)注聯(lián)合預(yù)測全景分割
論文/Paper: http://arxiv.org/pdf/2204.07157
代碼/Code: None
Cross-Image Relational Knowledge Distillation for Semantic Segmentation
標(biāo)題：語義分割的交叉圖像關(guān)系知識(shí)蒸餾
論文/Paper: http://arxiv.org/pdf/2204.06986
代碼/Code: https://github.com/winycg/cirkd
超分/Super-Resolution - 1 篇
Look Back and Forth: Video Super-Resolution with Explicit Temporal Difference Modeling
標(biāo)題：來回看：視頻超分辨率與明確的時(shí)間差異建模
論文/Paper: http://arxiv.org/pdf/2204.07114
代碼/Code: None
Transformers - - 2 篇
MiniViT: Compressing Vision Transformers with Weight Multiplexing
標(biāo)題：MinioIVIT：壓縮具有重量復(fù)用的視覺Transformers
論文/Paper: http://arxiv.org/pdf/2204.07154
代碼/Code: https://github.com/microsoft/cream
ViTOL: Vision Transformer for Weakly Supervised Object Localization
標(biāo)題：Vitol：視覺Transformers，用于弱監(jiān)督對(duì)象本地化
論文/Paper: http://arxiv.org/pdf/2204.06772
代碼/Code: https://github.com/Saurav-31/ViTOL
行人重識(shí)別/Person Re-Identification - 2 篇
Implicit Sample Extension for Unsupervised Person Re-Identification
標(biāo)題：無監(jiān)督行人重識(shí)別的隱式示例擴(kuò)展
論文/Paper: http://arxiv.org/pdf/2204.06892
代碼/Code: https://github.com/PaddlePaddle/PaddleClas
Clothes-Changing Person Re-identification with RGB Modality Only
標(biāo)題：改變的人只用RGB模態(tài)行人重識(shí)別
論文/Paper: http://arxiv.org/pdf/2204.06890
代碼/Code: https://github.com/guxinqian/Simple-CCReID.
其他/Other - 6 篇
What's in your hands? 3D Reconstruction of Generic Objects in Hands
標(biāo)題：你手中的是什么？三維重建在手中的通用物體
論文/Paper: http://arxiv.org/pdf/2204.07153
代碼/Code: None
GIFS: Neural Implicit Function for General Shape Representation
標(biāo)題：GIFS：一般形狀表示的神經(jīng)隱式功能
論文/Paper: http://arxiv.org/pdf/2204.07126
代碼/Code: None
The multi-modal universe of fast-fashion: the Visuelle 2.0 benchmark
標(biāo)題：fast-fashion的多模態(tài)宇宙：VISUELLE 2.0基準(zhǔn)
論文/Paper: http://arxiv.org/pdf/2204.06972
代碼/Code: None
Semi-Supervised Training to Improve Player and Ball Detection in Soccer
標(biāo)題：半監(jiān)督訓(xùn)練，以改善足球運(yùn)動(dòng)員和球偵查
論文/Paper: http://arxiv.org/pdf/2204.06859
代碼/Code: https://github.com/rvandeghen/SST
Pyramidal Attention for Saliency Detection
標(biāo)題：顯著性檢測的金字塔Attention
論文/Paper: http://arxiv.org/pdf/2204.06788
代碼/Code: https://github.com/tanveer-hussain/EfficientSOD2
OccAM's Laser: Occlusion-based Attribution Maps for 3D Object Detectors on LiDAR Data
標(biāo)題：Incam的激光：LIDAR數(shù)據(jù)上的3D對(duì)象探測器的基于遮擋的歸屬映射
論文/Paper: http://arxiv.org/pdf/2204.06577
代碼/Code: https://github.com/dschinagl/occam