Skip to content

Dataset · Hellfire98

SportsMOT sports video player tracking benchmark

Hellfire98/sportsmot

A large-scale video dataset in MOTChallenge format for multi-object tracking in dynamic sports scenes.

What's inside

The repository features the SportsMOT benchmark structured according to the MOTChallenge standard, with a total size of 1M to 10M records. The data consists of professional sports match video recordings annotated for player tracking. Scenes feature rapid motion with variable speed, frequent occlusions, motion blur, camera movement, and similar participant uniforms.

What you can do with it

  • Train and evaluate multi-object tracking (MOT) algorithms in complex dynamic conditions
  • Develop sports video analytics and automated player tracking systems
  • Test detector robustness to occlusion and object blur
Tasks
Object detection, Other
Languages
English
Size
1M–10M rows
Downloads this month
66.4K
Likes
1

How to load it

from datasets import load_dataset

ds = load_dataset("Hellfire98/sportsmot")