# FastSmoothSAM: A Fast Smooth Method For Segment Anything Model

arXiv: https://arxiv.org/abs/2507.15008 · HTML: https://mlsanity.com/paper/2507.15008/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 38 references that are in the map.
- First posted on arXiv 20 Jul 2025.

## Builds on (most cited first, 17 of 17)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [Holistically-Nested Edge Detection](https://mlsanity.com/paper/1504.06375/) (2015)
- [Segment Everything Everywhere All at Once](https://mlsanity.com/paper/2304.06718/) (2023)
- [YOLACT: Real-time Instance Segmentation](https://mlsanity.com/paper/1904.02689/) (2019)
- [Faster Segment Anything: Towards Lightweight SAM for Mobile Applications](https://mlsanity.com/paper/2306.14289/) (2023)
- [Deep Multi-modal Object Detection and Semantic Segmentation for Autonomous Driving: Datasets, Methods, and Challenges](https://mlsanity.com/paper/1902.07830/) (2019)
- [Fast Segment Anything](https://mlsanity.com/paper/2306.12156/) (2023)
- [EfficientSAM: Leveraged Masked Image Pretraining for Efficient Segment Anything](https://mlsanity.com/paper/2312.00863/) (2023)
- [SegViT: Semantic Segmentation with Plain Vision Transformers](https://mlsanity.com/paper/2210.05844/) (2022)
- [Semantic Segmentation using Vision Transformers: A survey](https://mlsanity.com/paper/2305.03273/) (2023)
- [Mask Transfiner for High-Quality Instance Segmentation](https://mlsanity.com/paper/2111.13673/) (2021)
- [MobileSAMv2: Faster Segment Anything to Everything](https://mlsanity.com/paper/2312.09579/) (2023)
- [AdaptiveSAM: Towards Efficient Tuning of SAM for Surgical Scene Segmentation](https://mlsanity.com/paper/2308.03726/) (2023)
- [EDTER: Edge Detection with Transformer](https://mlsanity.com/paper/2203.08566/) (2022)
- [SegViTv2: Exploring Efficient and Continual Semantic Segmentation with Plain Vision Transformers](https://mlsanity.com/paper/2306.06289/) (2023)

## Abstract (arXiv)

Accurately identifying and representing object edges is a challenging task in computer vision and image processing. The Segment Anything Model (SAM) has significantly influenced the field of image segmentation, but suffers from high memory consumption and long inference times, limiting its efficiency in real-time applications. To address these limitations, Fast Segment Anything (FastSAM) was proposed, achieving real-time segmentation. However, FastSAM often generates jagged edges that deviate from the true object shapes. Therefore, this paper introduces a novel refinement approach using B-Spline curve fitting techniques to enhance the edge quality in FastSAM. Leveraging the robust shape control and flexible geometric construction of B-Splines, a four-stage refining process involving two rounds of curve fitting is employed to effectively smooth jagged edges. This approach significantly improves the visual quality and analytical accuracy of object edges without compromising critical geometric information. The proposed method improves the practical utility of FastSAM by improving segmentation accuracy while maintaining real-time processing capabilities. This advancement unlocks greater potential for FastSAM technology in various real-world scenarios, such as industrial automation, medical imaging, and autonomous systems, where precise and efficient edge recognition is crucial.