Abstract: In recent years, open-vocabulary (OV) dense visual prediction (such as OV object detection, semantic, instance and panoptic segmentations) has attracted increasing research attention.
Abstract: Open-vocabulary semantic segmentation aims to assign pixel-level labels using arbitrary text queries, but existing CLIP-based methods often produce diffuse similarity maps and struggle with ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results