Multiple labels for same detection in Grounding DINO

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

Apache License 2.0

14.88k stars 1.38k forks source link

Multiple labels for same detection in Grounding DINO #428

Open Syzygianinfern0 opened 9 months ago

Syzygianinfern0 commented 9 months ago

Here is my text prompt.

TEXT_PROMPT = "Green Bowl . Red Bowl . Yellow Bowl . Blue Bowl . Green Block . Red Block . Yellow Block . Blue Block . Orange Block"

And here are the detections.

annotated_image

As you can see, the red block is labelled as "red block orange block". How can I prevent this and only have one label for each detection?

Syzygianinfern0 commented 9 months ago

@rentainhe I was wondering if you got to see this issue :)