以下CVPR2022論文打包合集：下載地址

神經網絡架構設計

[1] An Image Patch is a Wave: Quantum Inspired Vision MLP(圖像補丁是波浪：量子啟發的視覺 MLP)

paper | code | code

[2] A ConvNet for the 2020s

paper | code

解讀：“文藝復興” ConvNet卷土重來，壓過Transformer！FAIR重新設計純卷積新架構

三維視覺

[1] CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understanding(用于 3D 點云理解的自監督跨模態對比學習)

keywords: Self-Supervised Learning, Contrastive Learning, 3D Point Cloud, Representation Learning, Cross-Modal Learning

paper | code

[2] A Unified Query-based Paradigm for Point Cloud Understanding(一種基于統一查詢的點云理解范式)

paper

[3] X -Trans2Cap: Cross-Modal Knowledge Transfer using Transformer for 3D Dense Captioning(使用 Transformer 進行 3D 密集字幕的跨模式知識遷移)
keywords：Image Captioning and Dense Captioning(圖像字幕/密集字幕)；Knowledge distillation(知識蒸餾)；Transformer；3D Vision(三維視覺)

paper

[4] CLIP-NeRF: Text-and-Image Driven Manipulation of Neural Radiance Fields(文本和圖像驅動的神經輻射場操作)

keywords: NeRF, Image Generation and Manipulation, Language-Image Pre-Training (CLIP)

paper | code

姿態估計

[1] MixSTE: Seq2seq Mixed Spatio-Temporal Encoder for 3D Human Pose Estimation in Video(用于視頻中 3D 人體姿勢估計的 Seq2seq 混合時空編碼器)

keywords：3D Human Pose Estimation, Transformer

paper

[2] H4D: Human 4D Modeling by Learning Neural Compositional Representation(通過學習神經組合表示進行人體 4D 建模)

keywords: 4D Representation(4D 表征),Human Body Estimation(人體姿態估計),Fine-grained Human Reconstruction(細粒度人體重建)


paper

[3] Learning Local-Global Contextual Adaptation for Multi-Person Pose Estimation(學習用于多人姿勢估計的局部-全局上下文適應)

keywords:Top-Down Pose Estimation(從上至下姿態估計), Limb-based Grouping, Direct Regression


paper

圖像修復

[1] Incremental Transformer Structure Enhanced Image Inpainting with Masking Positional Encoding(增量transformer結構增強圖像修復與掩蔽位置編碼)

keywords: Image Inpainting, Transformer, Image Generation


paper | code

模型訓練

[1] DN-DETR: Accelerate DETR Training by Introducing Query DeNoising(通過引入查詢去噪加速 DETR 訓練)

keywords: Detection Transformer

paper | code

視覺語言表征學習

[1] HairCLIP: Design Your Hair by Text and Reference Image(通過文本和參考圖像設計你的頭發)

keywords: Language-Image Pre-Training (CLIP), Generative Adversarial Networks

paper | project

[2] Vision-Language Pre-Training with Triple Contrastive Learning(三重對比學習的視覺語言預訓練)

keywords: Vision-language representation learning, Contrastive Learning
paper | code

對比學習

[1] Crafting Better Contrastive Views for Siamese Representation Learning(為連體表示學習制作更好的對比視圖)

paper | code

深度估計

[1] OmniFusion: 360 Monocular Depth Estimation via Geometry-Aware Fusion(通過幾何感知融合進行 360 度單目深度估計)

keywords: monocular depth estimation(單目深度估計),transformer

paper

語義分割

[1] Class Re-Activation Maps for Weakly-Supervised Semantic Segmentation(弱監督語義分割的類重新激活圖)

paper | code

動作檢測

[1] Colar: Effective and Efficient Online Action Detection by Consulting Exemplars(通過咨詢示例進行有效且高效的在線動作檢測)

keywords:Online action detection(在線動作檢測)

paper

人臉偽造/反欺騙

[1] Protecting Celebrities with Identity Consistency Transformer(使用身份一致性transformer保護名人)

paper

長尾識別

[1] Targeted Supervised Contrastive Learning for Long-Tailed Recognition(用于長尾識別的有針對性的監督對比學習)

keywords: Long-Tailed Recognition(長尾識別), Contrastive Learning(對比學習)

paper

總結

以上是生活随笔為你收集整理的CVPR 2022 3月3日论文速递（19 篇打包下载）涵盖网络架构设计、姿态估计、三维视觉、动作检测、语义分割等方向的全部內容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網站內容還不錯，歡迎將生活随笔推薦給好友。