[論文筆記]CVPR2019_Learning Context Graph for Person Search
這篇論文是CVPR2019的oral，將graph learning引入了person search task中，文章很不錯，學習一下~ 1. Introduction 這篇文章對圖像中的context information（上下文信息）做了進一步的挖掘，利用其輔助person ...
原文：[論文筆記]CVPR2017_Joint Detection and Identification Feature Learning for Person Search
Title:Joint Detection and Identification Feature Learning for Person Search aXiv上該論文的第一個版本題目是 End to End Deep Learning for Person Search Authors:Tong Xiao Shuang Li Bochao Wang Liang Lin Xiaogang Wang ...
2019-07-03 13:03 0 1092 推薦指數：
相關推薦
[論文筆記]2014ICPR--Deep Metric Learning for Person Re-Identification
Title: Deep Metric Learning for Person Re-Identification Authors: Dong Yi, Zhen Lei, Shengcai Liao and Stan Z. Li Affiliation: Center ...
Person Re-identification 系列論文筆記（六）：AlignedReID
AlignedReID Zhang X, Luo H, Fan X, et al. AlignedReID: Surpassing Human-Level Performance in Person Re-Identification[J]. 2017. 本篇是來自曠視的一篇論文，剛發出來 ...
Person Re-identification 系列論文筆記（八）：SPReID
Human Semantic Parsing for Person Re-identification Kalayeh M M, Basaran E, Gokmen M, et al. Human Semantic Parsing for Person Re-identification ...
論文閱讀筆記（七十）【CVPR2021】：Combined Depth Space based Architecture Search For Person Re-identification
Introduction 現有主流方法采用了復雜的骨干網絡，參數量大，處理速度慢。因此本文的目標是構建一個計算效率更高、更適合ReID的輕量級網絡。 Neural Architecture Search(NAS)被利用來搜索輕量高效的網絡，但一般需要非常高昂的計算資源 ...
Person Re-identification 系列論文筆記（二）：A Discriminatively Learned CNN Embedding for Person Re-identification
A Discriminatively Learned CNN Embedding for Person Re-identification Zheng Z, Zheng L, Yang Y. A Discriminatively Learned CNN Embedding ...
CVPR19_Patch-based Discriminative Feature Learning for Unsupervised Person Re-identification
同樣來自中山大學Weishi Zheng老師組的一篇unsupervised person reid，但這篇不是domain adaption，而是只有unlabeled data。 PCB等監督模型的出現表明分塊對於行人重識別的有效性。因此這篇文章利用分塊的思想解決非監督問題 ...
論文閱讀筆記（二十五）【CVPR2020】：Weakly Supervised Discriminative Feature Learning with State Information for Person Identification
Introduction 每個行人圖片都會有視角、姿態等狀態信息，雖然自身不帶標簽，但可以預測這些狀態信息作為偽標簽，如下圖： 由於無監督學習通常存在結果出錯的問題，一些未標簽的樣本會偏離正確 ...