all-seeing icon indicating copy to clipboard operation
all-seeing copied to clipboard

Published 20 hours ago • verified publisher icon OpenGVLab

Where do the bounding boxes used in creating the AS-V2 dataset come from?

Open littlepenguin89106 opened this issue 5 months ago • 0 comments

Thank you for the excellent work on ASMv2. In the paper, you mention that when creating the AS-V2 dataset, the bounding boxes of objects are used as part of the prompt for GPT-4V. However, the process of obtaining these bounding boxes wasn't explained. Could you describe the workflow for acquiring the bounding boxes?

littlepenguin89106 avatar Sep 11 '24 01:09 littlepenguin89106