IoU: 81.0
Papers with Code · CC BY-SA 4.0
Referring Expression Segmentation
The task aims at labeling the pixels of an image or video that represent an object instance referred by a linguistic expression. In particular, the referring expression (RE) must allow the identification of an individual object in a discourse or scene (the referent). REs unambiguously identify the target instance.
논문
Overall IoU: 86.49Mean IoU: 86.64
Overall IoU: 84.2
IoU: 77.61IoU (%): 77.61
IoU: 80.0
Overall IoU: 85.7
Overall IoU: 80.64
Overall IoU: 85.3
Overall IoU: 83.18
IoU: 71.06
IoU: 80.6
Overall IoU: 82.8
Overall IoU: 78.24
Overall IoU: 83.6
Overall IoU: 77.83
Overall IoU: 79.7
Overall IoU: 69.79
Overall IoU: 82.19
Overall IoU: 81.74
mIoU: 79.64