IDEA-Research/T-Rex
A multimodal object detection model that synergizes text and visual prompts for open-set detection and counting tasks.
Not currently ranked — collecting fresh signals.
star history
T-Rex2 is a generic object detection framework that accepts both text and visual prompts to identify objects in images. The model enables interactive detection through visual exemplars and text descriptions, supporting open-set scenarios where classes are not pre-defined. It provides API access for integration and includes demo examples for easy experimentation.
Frequently asked
- What is IDEA-Research/T-Rex?
- A multimodal object detection model that synergizes text and visual prompts for open-set detection and counting tasks.
- Is T-Rex open source?
- Yes — IDEA-Research/T-Rex is an open-source project tracked on heatdrop.
- What language is T-Rex written in?
- IDEA-Research/T-Rex is primarily written in Python.
- How popular is T-Rex?
- IDEA-Research/T-Rex has 2.7k stars on GitHub.
- Where can I find T-Rex?
- IDEA-Research/T-Rex is on GitHub at https://github.com/IDEA-Research/T-Rex.