[paper note] Human-In-The-Loop Person Re-Identification
来源:互联网 发布:马蓉出轨内裤淘宝 编辑:程序博客网 时间:2024/05/07 17:17
- ECCV 2016
- Paper
- Poster
- Author: Hanxiao Wang, Shaogang Gong, Xiatian Zhu, Tao (Tony) Xiang
Professor Shaogang Gong from Queen Mary, University of London, who works closely with Dr. Tony Xiang, is an expert in person re-id field. He wrote a book naming Person Re-Identification. Since this book has been published, supervised learning method with CNN feature extractor has gradually dominate this field (person re-id). However, prof. Gong and his group are seeking for novel ways to resolve re-id problem. They have two papers about person re-id in ECCV 2016, together with Person Re-identification by Unsupervised L1 Graph Learning, both do not follow the supervised learning scheme.
Highlight
- Propose Human Verification Incremental Learning (HVIL), an online learning approach for person re-id.
- Do not need labeled data. Human participates in the training procedure, to give a pair of probe-gallery image a feedback as true, similar(but not true), dissimilar.
- Small training set, large test set.
- They show some other works attempting to relax the need of labeling, with semi-supervised, unsupervised and transfer learning approches, in Related Work part.
Modeling human feedback as a loss function
- Incrementally optimised ranking function
err(fxp(xg),y)=y(rank(fxp(xg)))
wherefxp(xg) is the distanse of pair{xp,xg} , which is defined as negative Mahalanobis Distance.y denotes the feedback, that is,y∈ {true-match, strong-negative, week-negative} ({m,s,w}).rank is just an int number denotes the rank of a gallery image. - Re-id ranking loss
y is defined asy(k)or=∑i=1kαiify∈{m,w}=∑i=k+1ngαiify∈{s}
withα1≥α2≥⋯≥0
Real-time Model Update for Instant Feedback Reward
- Negative Mahalanobis Distance:
fxp(xg)=−[(xp−xg)TM(xp−xg)],M∈Sd+ Sd+ represents semi-definate matrix. - Knowledge cumulation by online learning
Mt=argminM∈Sd+ΔF(M,Mt−1)+η(t)
This equation witht indicate that the matrixM in M-distance is learned in stages (knowledge cumulation).(t) is the loss of human feed back int stage.ΔF is Burg matrix divergence??ΔF(M,Mt−1)=tr(MM−1t−1)−logdet(MM−1t−1)
Metric ensemble learning
- When no human feedback is avilable.
- Idea: re-using pairs already verified by human
fensij=fensxpi(xgj)=−dTijWdij - Ideal ranking:
f∗ij=0 forci=cj andf∗ij=−1 forci≠cj .
Experiment
- Settings
- For human feedback, 300 people/image probe; 1000 people/image gallery. Return top-50 in the rank list for feedback.
- Max 3 rounds for each probe, result in 300-900 indicative verification.
- Claim that suer input will be 10-fold less.
- Better than other human-in-the-loop methods. Less feedback and search time.
- 56.1% on CUHK-03, 78% on Market-1501.
- Evaluate automated person re-id
- 168 pairs on CUHK-03, 234 pairs on Market-1501; supervised model trained with 300 ground truth data for comparison.
- Also compared with unensembled matrix after
τ (Mτ ) and average matrixM for all time1−τ (Mavg ) - Ensembled performs best.
0 0
- [paper note] Human-In-The-Loop Person Re-Identification
- [paper note] Human Re-identification in Crowd Videos using Personal, Social and Environmental (PSE)
- [paper note] Deep Attributes Driven Person Re-identification
- Re-ID:AlignedReID: Surpassing Human-Level Performance in Person Re-Identification 论文解析
- [Paper note] Embedding Deep Metric for Person Re-identification: A Study Against Large Variation
- [Paper note] Video-based Person Re-identification with Accumulative Motion Context
- [Paper note] MARS: A Video Benchmark for Large-Scale Person Re-identification
- [Paper note] Gated Siamese Convolutional Neural Network Architecture for Human Re-Identification
- 论文笔记之---Person Re-identification in the Wild
- 【Person Re-ID】In Defense of the Triplet Loss for Person Re-Identification
- ICCV2017: Unlabeled Samples Generated by GAN Improve the Person Re-Identification Baseline in Vitro
- 【论文笔记】In Defense of the Triplet Loss for Person Re-Identification
- Person Re-identification Datasets
- Person Re-identification Datasets
- Person Re-identification Overview
- Person Re-identification Overview
- Person Re-identification Datasets
- paper reading——《Improving Person Re-identification by Attribute and Identity Learning》
- Linux系统启动过程详解
- PHP中CURL方法curl_setopt()函数的参数
- hello bite.
- MP3格式音频文件结构解析
- JS判断移动端和PC端
- [paper note] Human-In-The-Loop Person Re-Identification
- mysql创建用户并授权
- 愁
- window文本和Linux文本的互相转换
- leetcode---Binary Search Tree Iterator
- lightoj1236(素因子分解)
- python 日期时间方法
- 采用集成学习算法提高分类器的准确性
- eclipse安装fatjar插件方法及版本不兼容问题解决