IDEA-Research / Grounded-Segment-Anything

Grounded SAM: Marrying Grounding DINO with Segment Anything & Stable Diffusion & Recognize Anything - Automatically Detect , Segment and Generate Anything

https://arxiv.org/abs/2401.14159

Can I use this on videos?

Jeredriq opened this issue 3 months ago · comments

Jeredriq commented 3 months ago

Ren Tianhe commented 3 months ago

Sure, you can use Grounded-SAM with Tracking model for video data

Jeredriq commented 3 months ago

@rentainhe thanks! Is there a google colab or huggingface page for this?

Ren Tianhe commented 3 months ago

@rentainhe thanks! Is there a google colab or huggingface page for this?

Yes, you can refer to https://github.com/hkchengrex/Tracking-Anything-with-DEVA for more details