Samreen Anjum and Danna Gurari
Presented at the Computer Vision for Microscopy Images (CVMI) workshop, CVPR 2020
This repository contains a Python implementation of the evaluation metric, TRA, inspired from the Cell Tracking Challenge and modified to support evaluation of tracking multiple objects with bounding boxes in videos.
We are hosting a challenge at the Computer Vision for Microscopy Images (CVMI) workshop at CVPR 2021. This challenge is hosted on the MOTChallenge platform. Each participant will have to create an account to submit their results.
Results Format:
Following the MOTChallenge guidelines, the file format should be the same as the ground truth file, which is a CSV text-file containing one object instance per line. Each line must contain 10 values:
<frame>, <id>, <bb_left>, <bb_top>, <bb_width>, <bb_height>, -1, -1, -1, -1
Evaluation:
The official code for evaluating the tracker is provided by the MOTChallenge team here. The script to run the evaluation code on the CTMC dataset can be found here.
If you have any questions or suggestions, please contact: Samreen Anjum - [email protected]