computer-vision-9

Multiple People Tracking

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in multiple-people-tracking-9

Trend

Dataset

Best Model

Actions

No benchmarks available.

Libraries

i

Use these libraries to find multiple-people-tracking-9 models and implementations

PaddlePaddle/PaddleDetection

2 papers 12,006

Datasets

MMPTRACK

LTFT

PersonPath22

Crowd in a rally | Crowd Counting | Crowd Human

Subtasks

No subtasks available.

Most implemented papers

MOT16: A Benchmark for Multi-Object Tracking

I. Reid, S. Roth, Anton Milan, L. Leal-Taixé, K. Schindler•Tue Mar 01 2016

A new release of the MOTChallenge benchmark, which focuses on multiple people tracking, and offers a significant increase in the number of labeled boxes, but also provides multiple object classes beside pedestrians and the level of visibility for every single object of interest.

1978

Content

bingfengyan/visam

2 papers 136

BingfengYan/CO-MOT

2 papers 54

0

Paper Graph

MOTRv2: Bootstrapping End-to-End Multi-Object Tracking by Pretrained Object Detectors

Tiancai Wang, Yuang Zhang, Xiangyu Zhang•Wed Nov 16 2022

This paper first adopt the anchor formulation of queries and then use an extra object detector to generate proposals as anchors, providing detection prior to MOTR, a simple yet effective pipeline to bootstrap end-to-end multi-object tracking with a pretrained object detector.

222 0

Paper Graph

Real-Time Multiple People Tracking with Deeply Learned Candidate Selection and Person Re-Identification

Long Chen, H. Ai, Zijie Zhuang, C. Shang•Sat Jun 30 2018

This paper proposes to handle unreliable detection by collecting candidates from outputs of both detection and tracking, and adopts a deeply learned appearance representation, which is trained on large-scale person re-identification datasets, to improve the identification ability of the tracker.

413 0

Paper Graph

The 1st-place Solution for ECCV 2022 Multiple People Tracking in Group Dance Challenge

Weiyao Lin, Tiancai Wang, Yuang Zhang, Xiangyu Zhang•Wed Oct 26 2022

This paper presents the 1st place solution to the Group Dance Multiple People Tracking Challenge, surpassing the second-place solution by +6.8% HOTA and using the YOLOX detection proposals for the anchor initialization of detect queries.

0 0

Paper Graph

Enhancing Multi-Camera People Tracking with Anchor-Guided Clustering and Spatio-Temporal Consistency ID Re-Assignment

Jenq-Neng Hwang, Hsiang-Wei Huang, Cheng-Yen Yang, Pyong-Kun Kim, Kwang-Ik Kim, Kyoungoh Lee, Zhongyu Jiang, Samartha Ramkumar, Chaitanya K. Mullapudi, In-Su Jang, Chung-I Huang•Tue Apr 18 2023

This work proposed a novel multi-camera multiple people tracking method that uses anchor-guided clustering for cross-camera re-identification and spatio-temporal consistency for geometry-based cross- camera ID reassigning that has demonstrated robustness and effectiveness in handling both synthetic and real-world data.

26 0

Paper Graph

MOT20: A benchmark for multi object tracking in crowded scenes

I. Reid, D. Cremers, S. Roth, Anton Milan, K. Schindler, Patrick Dendorfer, Hamid Rezatofighi, Javen Qinfeng Shi, L. Leal-Taix'e•Wed Mar 18 2020

The MOT20benchmark, consisting of 8 new sequences depicting very crowded challenging scenes, is presented, and gives to chance to evaluate state-of-the-art methods for multiple object tracking when handling extremely crowded scenarios.

774 0

Paper Graph

Do Different Tracking Tasks Require Different Appearance Models?

Philip H. S. Torr, Luca Bertinetto, Shengjin Wang, Yali Li, Hengshuang Zhao, Zhongdao Wang•Sun Jul 04 2021

UniTrack is presented, a solution to address five different tasks within the same framework that consists of a single and task-agnostic appearance model, which can be learned in a supervised or self-supervised fashion, and multiple ``heads'' that address individual tasks and do not require training.

100 0

Paper Graph

BoostTrack: boosting the similarity measure and detection confidence for improved multiple object tracking

Vukasin Stanojevic, Branimir T. Todorovic•Thu Apr 11 2024

The BoostTrack is presented, a simple yet effective tracing-by-detection MOT method that utilizes several lightweight plug and play additions to improve MOT performance, and a detection-tracklet confidence score is designed and used to scale the similarity measure and implicitly favour high detection confidence and high tracklet confidence pairs in one-stage association.

67 0

Paper Graph

Adding a benchmark result helps the community track progress.