GroundingDINO icon indicating copy to clipboard operation
GroundingDINO copied to clipboard

Error detecting multiple objects with single prompt

Open Saswati-Project opened this issue 5 months ago • 0 comments

I ran Grounding DINO on my own image, which contains boxes, an emergency alert, a lamp, and other objects. I used the official Colab notebook for Grounding DINO. According to the notebook, it's possible to detect multiple objects using a single prompt by separating them with commas.

However, when I try to detect three objects at once—like 'boxes, emergency alert, lamp'—none of them are detected. If I include only two objects, it detects only the last one.

Can you please help me understand why I'm unable to detect multiple objects using a single prompt?

Thank you in advance.

Saswati-Project avatar Sep 28 '25 20:09 Saswati-Project