Multi Object Tracking using Deep Learning

Multi Object Tracking is the task tracking objects in video frames. We survey the various types of MOT methods, with special focus on the latest methods that use Deep Learning.

Papers

• Learning to Track: Online Multi-object Tracking by Decision Making

  • ICCV'15
  • Yu Xiang, Alexandre Alahi, Silvio Savarese
  • cvgl.stanford.edu/papers/xiang_iccv15.pdf

• Online Multi-Target Tracking Using Recurrent Neural Networks

  • AAAI'17
  • Anton Milan, Seyed Hamid Rezatofighi, Anthony Dick, Ian Reid, Konrad Schindler
  • https://arxiv.org/abs/1604.03635

• Deep Network Flow for Multi-Object Tracking

  • CVPR ‘17
  • Samuel Schulter, Paul Vernaza, Wongun Choi, Manmohan Chandraker
  • https://arxiv.org/abs/1706.08482

• Learning to Track at 100 FPS with Deep Regression Networks

  • ECCV ‘16
  • https://link.springer.com/chapter/10.1007/978-3-319-46448-0_45

• Deep tracking in the wild: End-to-end tracking using recurrent neural networks

  • IJRR ‘17
  • http://www.robots.ox.ac.uk/~mobile/Papers/2017_IJRR_Dequaire.pdf
  • Julie Dequaire, Peter Ondrúška, Dushyant Rao, Dominic Wang, and Ingmar Posner

• Survey Slides by Merc Benz

  • https://32e624bb-a-62cb3a1a-s-sites.googlegroups.com/site/dlitsc17/A%20Survey%20on%20Leveraging%20Deep%20Neural%20Networks%20for.pdf

• Multiple Object Tracking: A Literature Review

  • https://arxiv.org/abs/1409.7618

• A survey on multiple object tracking algorithm

  • http://ieeexplore.ieee.org/document/7832121/

• Multiple Object Tracking: A Literature Review

  • https://pdfs.semanticscholar.org/8b42/8625c8a496c7e56b419197df55c751c22bc3.pdf

Students

• Vidit Jain
• Mohammed Sharfuddin
• Anjan Kumar
• Haard Panjal