Fast Segment Anything

By Xu Zhao, Wen-Yan Ding, Yongqi An, Yinglong Du, Tao Yu, Min Li, Ming Tang, Jinqiao Wang. arXiv.org, 2023

545 citations recorded by the source corpus.

The recently proposed segment anything model (SAM) has made a significant influence in many computer vision tasks. It is becoming a foundation step for many high-level tasks, like image segmentation, image caption, and image editing. However, its huge computation costs prevent it from wider applications in industry scenarios. The computation mainly comes from the Transformer architecture at high-resolution inputs. In this paper, we propose a speed-up alternative method for this fundamental task with comparable performance. By reformulating the task as segments-generation and prompting, we find that a regular CNN detector with an instance segmentation branch can also accomplish this task well. Specifically, we convert this task to the well-studied instance segmentation task and directly train the existing instance segmentation method using only 1/50 of the SA-1B dataset published by SAM authors. With our method, we achieve a comparable performance with the SAM method at 50 times higher run-time speed. We give sufficient experimental results to demonstrate its effectiveness. The codes and demos will be released at https://github.com/CASIA-IVA-Lab/FastSAM.

Subfields: Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/c01e94a36be5b3578fedb17205205b330290a778

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

1. Peng Gao
2. Hongsheng Li
3. Renrui Zhang
4. Ziyu Guo
5. Junting Pan
6. Xiaoliang Dai
7. Chaoning Zhang
8. Zhengkai Jiang

Citations in the corpus

• Foundation models in robotics: Applications, challenges, and the future (2023, unknown): “FastSAM [60] and MobileSAM [61] achieve comparable performance to SAM at faster inference speeds.”
• Faster Segment Anything: Towards Lightweight SAM for Mobile Applications (2023, result): “It is worth highlighting that our MobileSAM is 7 times smaller and 4 times faster than the concurrent FastSAM Zhao et al. [2023], while achieving superior performance.”
• EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything (2023, methodology): “MobileSAM[71] presents a decoupled distillation for obtaining a lightweight image encoder of SAM.”
• Personalize Segment Anything Model with One Shot (2023, background): “, 2023), faster inference speed (Zhao et al., 2023; Zhang et al., 2023a), all-purpose matching (Liu et al.”
• Foundation Models Defining a New Era in Vision: A Survey and Outlook (2025, unknown): “Similarly, visual prompt-based models like Segment Anything [11] would benefit from a real-time speed to ensure intractability [150], [192].”