Closed bhack closed 9 years ago
What do you mean by evaluation of cooperation? Are you involved in the GSoC 2013 OpenCV? I have no problem with cooperation for common OpenCV tracking API (that would be great), however, my performance evaluation efforts are currently focused on VOT challenge and the VOT evaluation kit.
Hi @lukacu , I have developed this tracking API for OpenCV library in the GSoC 2013 and I have ported two algorithms: MIL and Boosting. During the current GSoC another algorithm (LTD) is been integrated in the API. Also, the tracking API has a test framework based on Google Test that you can find in test folder of the module. It is partially based on this work [1]. These tests are very similar to VOT performance measures, experiment 1 is similar to OPE and experiment 2 is similar to SRE. Maybe we can evaluate a cooperation with a possible integration of test metodologies, what do you think about?
[1] Y. Wu, J. Lim, and M.-H. Yang, "Online object tracking: A benchmark," in Computer Vision and Pattern Recognition (CVPR), 2013, https://sites.google.com/site/trackerbenchmark/benchmarks/v10
Sure, that would be great. But I think we should get this discussion away from issue thread. You can find my contact information here: http://www.vicos.si/lukacu/ . As for the evaluation measures: there are some major differences of VOT in relation to [1] that we can also discuss.
This conversation was now moved to an email correspondence. We are discussing a possible collaboration between OpenCV and VOT. I am closing the issue.
https://github.com/Itseez/opencv_contrib/tree/master/modules/tracking