# TP-GMOT: Tracking Generic Multiple Object by Textual Prompt with Motion-Appearance Cost (MAC) SORT

Source: [TP-GMOT: Tracking Generic Multiple Object by Textual Prompt with Motion-Appearance Cost (MAC) SORT](<https://www.sota2.com/research/paper/68fa74ad1e209131cea63203>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-02T13:29:25.592Z

Authors: Duy Le Dinh Anh, Kim Hoang Tran, Ngan Hoang Le

Published: 2024-09-04T07:33:09.000Z

## Abstract

While Multi-Object Tracking (MOT) has made substantial advancements, it is limited by heavy reliance on prior knowledge and limited to predefined categories. In contrast, Generic Multiple Object Tracking (GMOT), tracking multiple objects with similar appearance, requires less prior information about the targets but faces challenges with variants like viewpoint, lighting, occlusion, and resolution. Our contributions commence with the introduction of the \\textbf{\\text{Refer-GMOT dataset}} a collection of videos, each accompanied by fine-grained textual descriptions of their attributes. Subsequently, we introduce a novel text prompt-based open-vocabulary GMOT framework, called \\textbf{\\text{TP-GMOT}}, which can track never-seen object categories with zero training examples. Within \\text{TP-GMOT} framework, we introduce two novel components: (i) {\\textbf{\\text{TP-OD}}, an object detection by a textual prompt}, for accurately detecting unseen objects with specific characteristics. (ii) Motion-Appearance Cost SORT \\textbf{\\text{MAC-SORT}}, a novel object association approach that adeptly integrates motion and appearance-based matching strategies to tackle the complex task of tracking multiple generic objects with high similarity. Our contributions are benchmarked on the \\text{Refer-GMOT} dataset for GMOT task. Additionally, to assess the generalizability of the proposed \\text{TP-GMOT} framework and the effectiveness of \\text{MAC-SORT} tracker, we conduct ablation studies on the DanceTrack and MOT20 datasets for the MOT task. Our dataset, code, and models will be publicly available at: https://fsoft-aic.github.io/TP-GMOT

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Generic Multiple Object Tracking on Refer-GMOT40](<https://www.sota2.com/research/sota/generic-multiple-object-tracking-on-refer-gmot40.md>): Generic Multiple Object Tracking; Refer-GMOT40; MOTA: 66.72.

- [Multiple Object Tracking on Refer-Animal](<https://www.sota2.com/research/sota/multiple-object-tracking-on-refer-animal.md>): Multiple Object Tracking; Refer-Animal; HOTA: 57.86.

- [Multi-Object Tracking on DanceTrack (val)](<https://www.sota2.com/research/sota/multi-object-tracking-on-dancetrack-val.md>): Multi-Object Tracking; DanceTrack (val); HOTA: 58.53.

- [Multiple Object Tracking on MOT20 (test)](<https://www.sota2.com/research/sota/multiple-object-tracking-on-mot20-test.md>): Multiple Object Tracking; MOT20 (test); IDF1: 79.2.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2409.02490>)

[PDF](<https://arxiv.org/pdf/2409.02490.pdf>)

[Implementation](<https://fsoft-aic.github.io/TP-GMOT>)