GroundingDINO
Visit ToolGroundingDINO is an open-source computer vision tool that performs open-set object detection. It combines DINO with grounded pre-training, allowing users to detect objects using natural language prompts.
At a glance
Trending
GroundingDINO is an open-source computer vision tool that performs open-set object detection. It combines DINO with grounded pre-training, allowing users to detect objects using natural language prompts.
Trending
About
GroundingDINO is an official implementation of the paper "Grounding DINO: Marrying DINO with Grounded Pre-Training for Open-Set Object Detection." This open-source tool provides PyTorch implementation and pre-trained models for advanced object detection. It excels in open-set detection, allowing users to identify objects using language descriptions, achieving high performance with COCO zero-shot 52.5 AP and COCO fine-tune 63.0 AP. GroundingDINO is also highly flexible, offering collaborations with Stable Diffusion and GLIGEN for image editing. The tool supports CPU-only mode, making it accessible on machines without GPUs, and provides various demos including Colab, Huggingface, and Gradio Web UI for ease of use.
Capabilities
Pricing & Plans
Open Source
Free
FAQs
Trending