# End-to-End Learning Deep CRF models for Multi-Object Tracking

arXiv: https://arxiv.org/abs/1907.12176 · HTML: https://mlsanity.com/paper/1907.12176/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jul 2019.

## Built on by (most cited first, 7 of 7)

- [Learning a Proposal Classifier for Multiple Object Tracking](https://mlsanity.com/paper/2103.07889/) (2021)
- [Recent Advances in Embedding Methods for Multi-Object Tracking: A Survey](https://mlsanity.com/paper/2205.10766/) (2022)
- [RoMo: A Robust Solver for Full-body Unlabeled Optical Motion Capture](https://mlsanity.com/paper/2410.02788/) (2024)
- [Interactive Multi-scale Fusion of 2D and 3D Features for Multi-object Tracking](https://mlsanity.com/paper/2203.16268/) (2022)
- [NOOUGAT: Towards Unified Online and Offline Multi-Object Tracking](https://mlsanity.com/paper/2509.02111/) (2025)
- [Joint Counting, Detection and Re-Identification for Multi-Object Tracking](https://mlsanity.com/paper/2212.05861/) (2022)
- [Enhancing Multi-Camera Gymnast Tracking Through Domain Knowledge Integration](https://mlsanity.com/paper/2511.16532/) (2025)

## Abstract (arXiv)

Existing deep multi-object tracking (MOT) approaches first learn a deep representation to describe target objects and then associate detection results by optimizing a linear assignment problem. Despite demonstrated successes, it is challenging to discriminate target objects under mutual occlusion or to reduce identity switches in crowded scenes. In this paper, we propose learning deep conditional random field (CRF) networks, aiming to model the assignment costs as unary potentials and the long-term dependencies among detection results as pairwise potentials. Specifically, we use a bidirectional long short-term memory (LSTM) network to encode the long-term dependencies. We pose the CRF inference as a recurrent neural network learning process using the standard gradient descent algorithm, where unary and pairwise potentials are jointly optimized in an end-to-end manner. Extensive experimental results on the challenging MOT datasets including MOT-2015 and MOT-2016, demonstrate that our approach achieves the state of the art performances in comparison with published works on both benchmarks.