TrackID3x3: A Dataset and Algorithm for Multi-Player Tracking with Identification and Pose Estimation in 3x3 Basketball Full-court Videos

2025-03-24

AI Summary

To address the lack of dedicated datasets and methods for multi-player tracking, identity recognition, and pose estimation in 3x3 basketball across diverse scenarios (indoor/outdoor fixed cameras and drone views), this work introduces the first comprehensive multi-task benchmark specifically designed for this sport, featuring fine-grained 2D pose annotations and cross-view ID-consistent labeling. We formally define a lightweight Track-ID joint task and propose an end-to-end framework integrating BoT-SORT with ReID for robust tracking and HRNet/RTMPose/SwinPose for accurate pose estimation. A specialized evaluation protocol is also introduced. Experiments demonstrate significant improvements in small-object ID stability and keypoint localization accuracy. The benchmark validates the robustness of mainstream models across three distinct subsets, establishing a reproducible paradigm and open-source benchmark for amateur-level, fixed-view sports analytics.

Technology Category

• Computer Vision: Motion & Tracking
• Intelligent Robots: State Estimation
• Machine Learning: Multi-instance/Multi-view Learning

Application Category

• Security and Privacy: Large-scale security measurements
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

Multi-object tracking, player identification, and pose estimation are fundamental components of sports analytics, essential for analyzing player movements, performance, and tactical strategies. However, existing datasets and methodologies primarily target mainstream team sports such as soccer and conventional 5-on-5 basketball, often overlooking scenarios involving fixed-camera setups commonly used at amateur levels, less mainstream sports, or datasets that explicitly incorporate pose annotations. In this paper, we propose the TrackID3x3 dataset, the first publicly available comprehensive dataset specifically designed for multi-player tracking, player identification, and pose estimation in 3x3 basketball scenarios. The dataset comprises three distinct subsets (Indoor fixed-camera, Outdoor fixed-camera, and Drone camera footage), capturing diverse full-court camera perspectives and environments. We also introduce the Track-ID task, a simplified variant of the game state reconstruction task that excludes field detection and focuses exclusively on fixed-camera scenarios. To evaluate performance, we propose a baseline algorithm called Track-ID algorithm, tailored to assess tracking and identification quality. Furthermore, our benchmark experiments, utilizing recent multi-object tracking algorithms (e.g., BoT-SORT-ReID) and top-down pose estimation methods (HRNet, RTMPose, and SwinPose), demonstrate robust results and highlight remaining challenges. Our dataset and evaluation benchmarks provide a solid foundation for advancing automated analytics in 3x3 basketball. Dataset and code will be available at https://github.com/open-starlab/TrackID3x3.

Problem

• Lack of datasets for 3x3 basketball tracking and pose estimation
• Need for player identification in fixed-camera sports scenarios
• Absence of benchmarks for multi-object tracking in amateur sports

Innovation

• Introduces TrackID3x3 dataset for 3x3 basketball
• Proposes Track-ID algorithm for tracking and identification
• Benchmarks with multi-object tracking and pose estimation