# Unveiling the Power of Self-supervision for Multi-view Multi-human Association and Tracking

arXiv: https://arxiv.org/abs/2401.17617 · HTML: https://mlsanity.com/paper/2401.17617/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 72 references that are in the map.
- First posted on arXiv 31 Jan 2024.

## Builds on (most cited first, 7 of 7)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [MOTChallenge 2015: Towards a Benchmark for Multi-Target Tracking](https://mlsanity.com/paper/1504.01942/) (2015)
- [Simple Unsupervised Multi-Object Tracking](https://mlsanity.com/paper/2006.02609/) (2020)
- [Self-supervised Learning for Video Correspondence Flow](https://mlsanity.com/paper/1905.00875/) (2019)
- [Locality Aware Appearance Metric for Multi-Target Multi-Camera Tracking](https://mlsanity.com/paper/1911.12037/) (2019)
- [Unsupervised Multiple Person Tracking using AutoEncoder-Based Lifted Multicuts](https://mlsanity.com/paper/2002.01192/) (2020)
- [Multiple Human Association between Top and Horizontal Views by Matching Subjects' Spatial Distributions](https://mlsanity.com/paper/1907.11458/) (2019)

## Built on by (most cited first, 1 of 1)

- [DOMR: Establishing Cross-View Segmentation via Dense Object Matching](https://mlsanity.com/paper/2508.04050/) (2025)

## Abstract (arXiv)

Multi-view multi-human association and tracking (MvMHAT), is a new but important problem for multi-person scene video surveillance, aiming to track a group of people over time in each view, as well as to identify the same person across different views at the same time, which is different from previous MOT and multi-camera MOT tasks only considering the over-time human tracking. This way, the videos for MvMHAT require more complex annotations while containing more information for self learning. In this work, we tackle this problem with a self-supervised learning aware end-to-end network. Specifically, we propose to take advantage of the spatial-temporal self-consistency rationale by considering three properties of reflexivity, symmetry and transitivity. Besides the reflexivity property that naturally holds, we design the self-supervised learning losses based on the properties of symmetry and transitivity, for both appearance feature learning and assignment matrix optimization, to associate the multiple humans over time and across views. Furthermore, to promote the research on MvMHAT, we build two new large-scale benchmarks for the network training and testing of different algorithms. Extensive experiments on the proposed benchmarks verify the effectiveness of our method. We have released the benchmark and code to the public.