We provide a new Synchronized Large-scale Person re-identification dataset called SLP constructed for practical large-scale surveillance scenarios.

The main characteristics of SLP :

• A large number of people: The total number of people in the dataset is 2,632. Among them, 1,737 people appear in at least two cameras.
• Nine cameras: The dataset was captured from nine non-overlapping cameras.
• Multiple appearances availability: It provides multiple appearances of each person. In this version, we do not provide the original image appearances but provide pre-computed feature descriptors only.
• Scene background availability: It provides the scene background. Although it only contains the background of the camera scene, it contains scene structures such as buildings, ground plane, orthogonal lines etc., which are quite useful for understanding the scene.
• People trajectory availability: The tracking information of every person is available. It includes positions (x,y locations) and sizes (height, width) of people in entire image sequences.
• Camera synchronization: Each camera is synchronized with a global time-stamp. It provides all time-stamps of video frames.
• Camera calibration: Each camera is calibrated and camera parameters are given.

When you use SLP dataset please cite below papers

[1] Yeong-Jun Cho, Su-A kim*, Jae-Han Park*, Kyuewang Lee and Kuk-Jin Yoon "Joint Person Re-identification and Camera Network Topology Inference in Multiple Cameras", Arxiv: 1710.00983

[2] Yeong-Jun Cho, Jae-Han Park*, Su-A Kim*, Kyuewang Lee and Kuk-Jin Yoon "Unified Framework for Automated Person Re-identification and Camera Network Topology Inference in Camera Networks", International Workshop on Cross-domain Human Identification (in conjunction with ICCV), 2017

THIS DATASET CANNOT BE USED FOR COMMERCIAL USE.

✉ engzenia@gmail.com