Referring Image Segmentation (RIS) is a fundamental vision-language task that outputs object masks based on text descriptions. Many works have achieved considerable progress for RIS, including ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results