Towards a Principled Integration of Multi-Camera Re-Identification and Tracking through Optimal Bayes Filters
arXiv:1705.04608
Abstract
With the rise of end-to-end learning through deep learning, person detectors and re-identification (ReID) models have recently become very strong. Multi-camera multi-target (MCMT) tracking has not fully gone through this transformation yet. We intend to take another step in this direction by presenting a theoretically principled way of integrating ReID with tracking formulated as an optimal Bayes filter. This conveniently side-steps the need for data-association and opens up a direct path from full images to the core of the tracker. While the results are still sub-par, we believe that this new, tight integration opens many interesting research opportunities and leads the way towards full end-to-end tracking from raw pixels.
arXiv commentFirst two authors have equal contribution. This is initial work into a new direction, not a benchmark-beating method. v2 only adds acknowledgements and fixes a typo in e-mail
References in corpus (9)
- In Defense of the Triplet Loss for Person Re-Identification
- MOT16: A Benchmark for Multi-Object Tracking
- Person Re-identification: Past, Present and Future
- A Discriminatively Learned CNN Embedding for Person Re-identification
- MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking
- Transferring Rich Feature Hierarchies for Robust Visual Tracking
- Deep Transfer Learning for Person Re-identification
- The STRANDS Project: Long-Term Autonomy in Everyday Environments
- Tracking The Untrackable: Learning To Track Multiple Cues with Long-Term Dependencies