List your company
ResearchFormerly WizWand

Multiple Object Grounding on Reasoning Segmentation Long query

49.2gIoULISA
21.1228.4135.742.99May 29, 2025
Updated 8mo ago

Evaluation results

MethodLinks
LISATraining paradigm=Train LLaVA 1.5-7B, Query source=Original2025.0549.248.9
TextRegionQuery source=Interpreted from LLaVA 1.5-7B2025.0547.245.3
OVSegQuery source=Interpreted from LLaVA 1.5-7B2025.0544.637.1
TridentQuery source=Interpreted from LLaVA 1.5-7B2025.0543.342.5
OVSegTraining paradigm=Training-based, VLM usage=No, Query source=Original2025.0528.722.5
TextRegionTraining paradigm=Training-free, VLM usage=No, Query source=Original2025.0526.424.1
SEEMTraining paradigm=Training-based, VLM usage=No, Query source=Original2025.0525.620.8
TridentTraining paradigm=Training-free, VLM usage=No, Query source=Original2025.0524.523.1
Grounded-SAMTraining paradigm=Training-based, VLM usage=No, Query source=Original2025.0522.418.6
X-DecoderTraining paradigm=Training-based, VLM usage=No, Query source=Original2025.0522.217.5