[NEW] The DukeMTMC dataset is taken down due to privacy concerns... Will reopen when it's clear.

[NEW] Benchmark evalution code-base is released! Please refer to this repo for more details.

DukeMTMC4ReID

DukeMTMC4ReID dataset is new large-scale real-world person re-id dataset based on DukeMTMC. We use a fast state-of-the-art person detector for accurate detections. After verified by the ground truth, for each identity, we uniformly sample 5 "good" bounding boxes in each available camera, while retaining all the "FP" bounding boxes in the corresponding frames. To summarize, the relevant statistics of the proposed DukeMTMC4ReID dataset are provided below:

• Images corresponding to 1,852 people existing across all the 8 cameras
• 1,413 unique identities with 22,515 bounding boxes that appear in more than one camera (valid identities)
• 439 distractor identities with 2,195 bounding boxes that appear in only one camera, in addition to 21,551 ?FP? bounding boxes from the person detector
• The size of the bounding box varies from 72×34 pixels to 415×188 pixels

More details and benchmark results can be found in this paper

How to use

1. Clone or download this repo

2. Download the dataset from here and extract it within the same folder of the code

   • p0001_c5_f0000051246_1.jpg
     bounding box of person 0001 in camera 5 at frame 51246
   • partition.
     idx_train - index of train samples
     idx_test - index of test samples
     idx_probe - index of probe samples in test
     idx_gallery - index of gallery samples in test
     ix_pos_pair - index of pre-generated positive pairs
     ix_neg_pair - index of pre-generated negtive pairs
     cam_pairs - [probe camera, gallery camera] (0 means all the other cameras)

3. Download the pre-computed feature

4. run script_test.m to parsing the data and evaluate it with pre-computed feature

References

@InProceedings{gou2017dukemtmc4reid,
  author = {Gou, Mengran and Karanam, Srikrishna and Liu, Wenqian and Camps, Octavia and Radke, Richard J.},
  title = {DukeMTMC4ReID: A Large-Scale Multi-Camera Person Re-Identification Dataset},
  booktitle = {The IEEE Conference on Computer Vision and Pattern Recognition (CVPR) Workshops},
  month = {July},
  year = {2017}
}

If you use this dataset, please also cite the original DukeMTMC dataset accordingly:

@inproceedings{ristani2016MTMC,
  title = {Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking},
  author = {Ristani, Ergys and Solera, Francesco and Zou, Roger and Cucchiara, Rita and Tomasi, Carlo},
  booktitle = {European Conference on Computer Vision workshop on Benchmarking Multi-Target Tracking},
  year = {2016}
}

License

Please refer to the license file for DukeMTMC4ReID and DukeMTMC