DukeMTMC aims to accelerate advances in multi-target multi-camera tracking. It provides a tracking system that works within and across cameras, a new large scale HD video data set recorded by 8 synchronized cameras with more than 7,000 single camera trajectories and over 2,000 unique identities, and a new performance evaluation method that measures how often a system is correct about who is where.
DukeMTMC is a new, manually annotated, calibrated, multi-camera data set recorded outdoors on the Duke University campus with 8 synchronized cameras. It consists of:
New! A faster version of the evaluation kit is now available.
Click here for downloads and more information.
New! Below is a list of dataset extensions provided by the community:
If you use or extend DukeMTMC, please refer to the license terms.
For the up-to-date scoreboard visit motchallenge.net. Here you will find the official mtmc-devkit used for evaluation by MOTChallenge. For instructions how to submit on motchallenge you can refer to this link.
Our identity-based measures compute how often the system is correct about who is where, regardless of how often a target is lost and reacquired. Our measures are useful in applications such as security, surveillance or sports. We provide code for a small demo that implements our evaluation. The details of our measures appear in our paper.
We provide code for the following tracking systems which are all based on Correlation Clustering optimization:
 Performance Measures and a Data Set for Multi-Target, Multi-Camera Tracking. E. Ristani, F. Solera, R. S. Zou, R. Cucchiara and C. Tomasi. ECCV 2016 Workshop on Benchmarking Multi-Target Tracking. [pdf]
 Tracking Social Groups Within and Across Cameras. F. Solera, S. Calderara, E. Ristani, C. Tomasi, and R. Cucchiara. IEEE Transactions on Circuits and Systems 2016. [pdf]
 Tracking Multiple People Online and in Real Time. E. Ristani and C. Tomasi. ACCV 2014. [pdf]
If you use our work, please cite our papers accordingly:
If you use or extend our data, please see the license terms.