om-ai-lab/OmDet

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1393 stars · 118 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

OmDet-Turbo is a PyTorch implementation of a transformer-based open-vocabulary object detection model that detects arbitrary user-defined object categories in real time. It achieves state-of-the-art zero-shot detection performance with over 100 FPS inference on an A100 GPU and is integrated into Hugging Face Transformers.

Use cases

• detect arbitrary objects in images without retraining
• run real-time open-vocabulary object detection on video streams
• perform zero-shot detection on custom object categories
• export detection model to ONNX for deployment
• evaluate open-vocabulary detection on COCO, LVIS, ODinW, and OVDEval

When to choose

• you need fast real-time detection of object categories not in a fixed training vocabulary
• you want zero-shot detection with strong generalization to new domains
• you need a model already integrated with Hugging Face Transformers

When to avoid

• you only need detection of a fixed set of classes where a classic detector like YOLO suffices
• you lack GPU hardware for inference
• you need segmentation or other vision tasks beyond bounding-box detection

Facets

library · maturity active

computer-vision image-processing machine-learning deep-learning computer-vision machine-learning artificial-intelligence python object-detection open-vocabulary-detection zero-shot-detection vision-language real-time-inference pytorch onnx-export gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="om-ai-lab/OmDet")