Archive
Multimodal & Vision

Three Levels Of Grounding

Archive
Expert
300pts0 solves
Open-vocabulary grounding research distinguishes three prediction-target granularities: _____(1) box, pixel _____(2), and _____(3) detection. Fill the 3 blanks. Flag format: CONGRESS{1:[word],2:[word],3:[word]}. Example: CONGRESS{1:region,2:mask,3:keypoint}.
Show hint
Box, shape, point.

Archive — no submissions accepted

This challenge is preserved for reference. Play live challenges at /challenges.