MasterBin-IIAU/UNINEXT
A unified deep learning framework for object detection, tracking, instance segmentation, and referring expression comprehension across images and video.
Not currently ranked — collecting fresh signals.
star history
UNINEXT is a CVPR 2023 paper implementation that unifies multiple computer vision tasks into a single model, handling object discovery and retrieval across diverse perception tasks. It supports instance segmentation, multi-object tracking, video object segmentation, and referring expression comprehension. The framework aims to treat different instance perception tasks as a unified object discovery and retrieval problem.
Frequently asked
- What is MasterBin-IIAU/UNINEXT?
- A unified deep learning framework for object detection, tracking, instance segmentation, and referring expression comprehension across images and video.
- Is UNINEXT open source?
- Yes — MasterBin-IIAU/UNINEXT is open source, released under the MIT license.
- What language is UNINEXT written in?
- MasterBin-IIAU/UNINEXT is primarily written in Python.
- How popular is UNINEXT?
- MasterBin-IIAU/UNINEXT has 1.3k stars on GitHub.
- Where can I find UNINEXT?
- MasterBin-IIAU/UNINEXT is on GitHub at https://github.com/MasterBin-IIAU/UNINEXT.