Welcome to the 9th ABC Challenge
Brought to you by Nagoya University in conjunction with ABC

Overview

TrackID3x3 is a fixed-camera basketball analysis dataset and challenge designed for multi-player tracking and identification in 3x3 basketball. The system focuses on detecting all six on-court players, estimating their positions on a standardized court, maintaining consistent tracks over time, and identifying each player using team affiliation and jersey number. The challenge evaluates multi-player tracking and identification using the official Training set in the top view above and the released oblique-view Challenge videos.

Important Dates and Submissions

• Dataset including Challenge data: Already available

• Results submission: Google Form (in preparation) by November 3, 2026, 23:59 AoE（UTC−12）

• Paper and code submission: CMT (in preparation) by  November 10, 2026