hnhbcc / person-re-ranking

Person Re-ranking (CVPR 2017)

Re-ranking Person Re-identification with k-reciprocal Encoding

================================================================

This code has the source code for the paper "Re-ranking Person Re-identification with k-reciprocal Encoding". Including:

1. IDE baseline
2. Re-ranking code
3. CUHK03 new training/testing protocol

If you find this code useful in your research, please consider citing:

@article{zhong2017re,
  title={Re-ranking Person Re-identification with k-reciprocal Encoding},
  author={Zhong, Zhun and Zheng, Liang and Cao, Donglin and Li, Shaozi},
  booktitle={CVPR},
  year={2017}
}

The neighbor encoding method of our paper is inspired by the reference [2]. If you use the re-ranking code in your paper, please also cite:

@article{bai2016sparse,
  title={Sparse contextual activation for efficient visual re-ranking},
  author={Bai, Song and Bai, Xiang},
  journal={IEEE Transactions on Image Processing},
  year={2016},
  publisher={IEEE}
}

================================================================

The new training/testing protocol for CUHK03

The new protocol splits the CUHK03 dataset into training set and testing set similar to that of Market-1501, which consist of 767 identities and 700 identities respectively.

In testing, we randomly select one image from each camera as the query for each identity and use the rest of images to construct the gallery set. We make sure that each query identity is selected by both two cameras, so that cross-camera search can be performed.

In evaluation, true matched images captured in the same camera as the query are viewed as “junk”. Meaning that junk images is of zero influence to re-id accuracy (CMC/mAP).

The new training/testing protocol split for CUHK03 in our paper is in the "evaluation/data/CUHK03/" folder.

• cuhk03_new_protocol_config_detected.mat
• cuhk03_new_protocol_config_labeled.mat

	Labeled	detected
#Training	7,368	7,365
#Query	1,400	1,400
#Gallery	5,328	5,332

State-of-the-art

-	Labeled	Labeled	detected	detected
Methods	Rank@1	mAP	Rank@1	mAP	Reference
BOW+XQDA	7.93%	7.29%	6.36%	6.39%	"Scalable person re-identification: a benchmark", Zheng Liang, Shen Liyue, Tian Lu, Wang Shengjin, Wang Jingdong and Tian, Qi, ICCV 2015 [project]
PUL	-	-	9.1%	9.2%	"Unsupervised Person Re-identification: Clustering and Fine-tuning", Hehe Fan, Liang Zheng and Yi Yang, arXiv:1705.10444 [code]
LOMO+XQDA	14.8%	13.6%	12.8%	11.5%	"Person Re-identification by Local Maximal Occurrence Representation and Metric Learning", Liao Shengcai, Hu Yang, Zhu Xiangyu and Li Stan Z, CVPR 2015 [project]
IDE	22.2%	21.0%	21.3%	19.7%	"Person Re-identification: Past, Present and Future", Zheng Liang, Yi Yang, and Alexander G. Hauptmann, arXiv:1610.02984 [code]
IDE+XQ.+re-ranking	38.1%	40.3%	34.7%	37.4%	"Re-ranking Person Re-identification with k-reciprocal Encoding", Zhun Zhong, Liang Zheng, Donglin Cao,Shaozi Li, CVPR 2017 [code]
PAN	36.9%	35.0%	36.3%	34.0%	"Pedestrian Alignment Network for Person Re-identification", Zhedong Zheng, Liang Zheng, Yi Yang, arXiv:1707.00408 [code]
PAN+re-ranking	43.9%	45.8%	41.9%	43.8%	"Pedestrian Alignment Network for Person Re-identification", Zhedong Zheng, Liang Zheng, Yi Yang, arXiv:1707.00408 [code]
SVDNet	40.93	37.83	41.5%	37.26%	"SVDNet for Pedestrian Retrieval", Sun Yifan, Zheng Liang, Deng Weijian, Wang Shengjin, ICCV 2017
SVDNet+re-ranking	47.93	50.05	46.43%	48.95%	"SVDNet for Pedestrian Retrieval", Sun Yifan, Zheng Liang, Deng Weijian, Wang Shengjin, ICCV 2017
SVDNet+XQDA	44.07	40.06	42.71%	39.08%	"SVDNet for Pedestrian Retrieval", Sun Yifan, Zheng Liang, Deng Weijian, Wang Shengjin, ICCV 2017
SVDNet+XQDA+re-ranking	49.86	51.83	48.7%	50.2%	"SVDNet for Pedestrian Retrieval", Sun Yifan, Zheng Liang, Deng Weijian, Wang Shengjin, ICCV 2017

================================================================

IDE Baseline + Re-ranking

Requirements: Caffe

Requirements for Caffe and matcaffe (see: Caffe installation instructions)

Installation

1. Build Caffe and matcaffe

   cd $Re-ranking_ROOT/caffe
   # Now follow the Caffe installation instructions here:
   # http://caffe.berkeleyvision.org/installation.html
   make -j8 && make matcaffe

2. Download pre-computed imagenet models, Market-1501 dataset and CUHK03 dataset

Please download the pre-trained imagenet models and put it in the "data/imagenet_models" folder.
Please download Market-1501 dataset and unzip it in the "evaluation/data/Market-1501" folder. 
Please download CUHK03 dataset and unzip it in the "evaluation/data/CUHK03" folder.

• Pre-trained imagenet models

• Market-1501

• CUHK03 [Baiduyun] [Google drive]

Training and testing IDE model

1. Training

cd $Re-ranking_ROOT
# train IDE ResNet_50 for Market-1501
./experiments/Market-1501/train_IDE_ResNet_50.sh

# train IDE ResNet_50 for CUHK03
./experiments/CUHK03/train_IDE_ResNet_50_labeled.sh
./experiments/CUHK03/train_IDE_ResNet_50_detected.sh

2. Feature Extraction

cd $Re-ranking_ROOT/evaluation
# extract feature for Market-1501
matlab Market_1501_extract_feature.m

# extract feature for CUHK03
matlab CUHK03_extract_feature.m

3. Evaluation

# evaluation for Market-1501
matlab Market_1501_evaluation.m
  
# evaluation for CUHK03
matlab CUHK03_evaluation.m

Results

You can download our pre-trained IDE models and IDE features, and put them in the "output" and "evaluation/feat" folder, respectively.

• IDE models [Baiduyun] [Google drive]

• IDE features [Baiduyun] [Google drive]

Using the above IDE models and IDE features, you can reproduce the results with our re-ranking method as follows:

• Market-1501

Methods	Rank@1	mAP
IDE_ResNet_50 + Euclidean	78.92%	55.03%
IDE_ResNet_50 + Euclidean + re-ranking	81.44%	70.39%
IDE_ResNet_50 + XQDA	77.58%	56.06%
IDE_ResNet_50 + XQDA + re-ranking	80.70%	69.98%

For Market-1501, these results are better than those reported in our paper, since we add a dropout = 0.5 layer after pool5.

• CUHK03 under the new training/testing protocol

	Labeled	Labeled	detected	detected
Methods	Rank@1	mAP	Rank@1	mAP
BOW + XQDA [1]	7.93%	7.29%	6.36%	6.39%
BOW + XQDA + re-ranking	8.93%	9.94%	8.29%	8.81%
LOMO + XQDA [3]	14.8%	13.6%	12.8%	11.5%
LOMO + XQDA + re-ranking	19.1%	20.8%	16.6%	17.8%
IDE_CaffeNet + Euclidean	15.6%	14.9%	15.1%	14.2%
IDE_CaffeNet + Euclidean + re-ranking	19.1%	21.3%	19.3%	20.6%
IDE_CaffeNet + XQDA	21.9%	20.0%	21.1%	19.0%
IDE_CaffeNet + XQDA + re-ranking	25.9%	27.8%	26.4%	26.9%
IDE_ResNet_50 + Euclidean	22.2%	21.0%	21.3%	19.7%
IDE_ResNet_50 + Euclidean + re-ranking	26.6%	28.9%	24.9%	27.3%
IDE_ResNet_50 + XQDA	32.0%	29.6%	31.1%	28.2%
IDE_ResNet_50 + XQDA + re-ranking	38.1%	40.3%	34.7%	37.4%

References

[1] Scalable Person Re-identification: A Benchmark. Zheng, Liang and Shen, Liyue and Tian, Lu and Wang, Shengjin and Wang, Jingdong and Tian, Qi. In ICCV 2015.

[2] Sparse contextual activation for efficient visual re-ranking. Bai, Song and Bai, Xiang. IEEE Transactions on Image Processing. 2016

[3] Person re-identification by local maximal occurrence representation and metric learning. Liao S, Hu Y, Zhu X, et al. In CVPR. 2015

Contact us

If you have any questions about this code, please do not hesitate to contact us.

Zhun Zhong

Liang Zheng