In this paper, we introduce Semantic-SAM, a universal image segmentation model to enable segment and recognize anything at any desired granularity. Our model offers two key advantages: semantic-awareness and granularity-abundance. To achieve semantic-awareness, we consolidate multiple datasets across three granularities and introduce decoupled classification for objects and parts. This allows our model to capture rich semantic information. For the multi-granularity capability, we propose a multi-choice learning scheme during training, enabling each click to generate masks at multiple levels that correspond to multiple ground-truth masks. Notably, this work represents the first attempt to jointly train a model on SA-1B, generic, and part segmentation datasets. Experimental results and visualizations demonstrate that our model successfully achieves semantic-awareness and granularity-abundance. Furthermore, combining SA-1B training with other segmentation tasks, such as panoptic and part segmentation, leads to performance improvements. We will provide code and a demo for further exploration and evaluation.
@article{arxiv.2307.04767,
title = {Semantic-SAM: Segment and Recognize Anything at Any Granularity},
author = {Feng Li and Hao Zhang and Peize Sun and Xueyan Zou and Shilong Liu and Jianwei Yang and Chunyuan Li and Lei Zhang and Jianfeng Gao},
journal= {arXiv preprint arXiv:2307.04767},
year = {2023}
}