Improving Generalized Visual Grounding with Instance-aware Joint Learning
Open paper
Benchmark
Model
Rank
Results
generalized-referring-expression-comprehension-on-grefcoco
InstanceVG
#1
Precision@(F1=1, IoU≥0.5): 73.5
N-acc.: 72.8
generalized-referring-expression-segmentation-on-grefcoco
InstanceVG
#2
gIoU: 73.36
cIoU: 69.22
Rank counts only results with a code link.