HyperSeg: Towards Universal Visual Segmentation with Large Language Model

Benchmark Model Rank Results
open-vocabulary-semantic-segmentation-on-1HyperSeg#1mIoU: 64.6
open-vocabulary-semantic-segmentation-on-5HyperSeg#10mIoU: 92.1
panoptic-segmentation-on-coco-minivalHyperSeg (Swin-B)#1PQ: 61.2
referring-expression-segmentation-on-davisHyperSeg#2J&F 1st frame: 71.2
referring-expression-segmentation-on-refcocoHyperSeg#2Overall IoU: 84.8
referring-expression-segmentation-on-refcoco-3HyperSeg#3Overall IoU: 79.0
referring-expression-segmentation-on-refcoco-4HyperSeg#1Overall IoU: 83.5
referring-expression-segmentation-on-refcoco-5HyperSeg#2Overall IoU: 75.2
referring-expression-segmentation-on-refcoco-8HyperSeg#2Overall IoU: 85.7
referring-expression-segmentation-on-refcoco-9HyperSeg#1Overall IoU: 83.4
referring-expression-segmentation-on-refcocogHyperSeg#2Overall IoU: 79.4
referring-expression-segmentation-on-refcocog-1HyperSeg#4Overall IoU: 78.9
referring-video-object-segmentation-on-referHyperSeg#3J&F: 68.5
semantic-segmentation-on-coco-1HyperSeg#1mIoU: 77.2