visionml/pytracking
A PyTorch framework implementing deep learning-based visual trackers including DiMP, ATOM, and other models with complete training code.
Not currently ranked — collecting fresh signals.
star history
This library provides implementations of various visual object tracking algorithms and video object segmentation methods using PyTorch. It includes trained models for multiple trackers (TaMOs, RTS, ToMP, KeepTrack, DiMP, ATOM) along with a complete training framework called LTR for learning tracking representations. The project offers evaluation tools, benchmark datasets, and model checkpoints for tracking research.
Frequently asked
- What is visionml/pytracking?
- A PyTorch framework implementing deep learning-based visual trackers including DiMP, ATOM, and other models with complete training code.
- Is pytracking open source?
- Yes — visionml/pytracking is open source, released under the GPL-3.0 license.
- What language is pytracking written in?
- visionml/pytracking is primarily written in Python.
- How popular is pytracking?
- visionml/pytracking has 3.5k stars on GitHub.
- Where can I find pytracking?
- visionml/pytracking is on GitHub at https://github.com/visionml/pytracking.