tiassap/object-tracking_101

Basic implementation of object tracking algorithm

object-tracking_101

Basic implementation of object tracking algorithm.

Object tracking with SORT algorithm on MOT-15. Faster R-CNN is used as object detection algorithm. Still have some problems with occlusion here. (Short term occlusion is okay). See video

Run this command line for real-time inference using video camera. Note: press Esc to exit.

$ python main.py 

For inference on video:

$ python main.py --video <path/to/video>

To save output video (use arg --video <path/to/video> to use video file):

$ python main.py --save

1. SORT (Simple Online and Realtime Tracking)

@inproceedings{Bewley2016_sort,
  author={Bewley, Alex and Ge, Zongyuan and Ott, Lionel and Ramos, Fabio and Upcroft, Ben},
  booktitle={2016 IEEE International Conference on Image Processing (ICIP)},
  title={Simple online and realtime tracking},
  year={2016},
  pages={3464-3468},
  keywords={Benchmark testing;Complexity theory;Detectors;Kalman filters;Target tracking;Visualization;Computer Vision;Data Association;Detection;Multiple Object Tracking},
  doi={10.1109/ICIP.2016.7533003}
}