# Semantic-Fast-SAM: Efficient Semantic Segmenter

Source: [Semantic-Fast-SAM: Efficient Semantic Segmenter](<https://www.sota2.com/research/paper/69e9a7ab55a3c32af89f715e>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-01T05:00:18.486Z

Authors: Byunghyun Kim

Published: 2026-04-22T04:18:39.000Z

## Abstract

We propose Semantic-Fast-SAM (SFS), a semantic segmentation framework that combines the Fast Segment Anything model with a semantic labeling pipeline to achieve real-time performance without sacrificing accuracy. FastSAM is an efficient CNN-based re-implementation of the Segment Anything Model (SAM) that runs much faster than the original transformer-based SAM. Building upon FastSAM's rapid mask generation, we integrate a Semantic-Segment-Anything (SSA) labeling strategy to assign meaningful categories to each mask. The resulting SFS model produces high-quality semantic segmentation maps at a fraction of the computational cost and memory footprint of the original SAM-based approach. Experiments on Cityscapes and ADE20K benchmarks demonstrate that SFS matches the accuracy of prior SAM-based methods (mIoU ~ 70.33 on Cityscapes and 48.01 on ADE20K) while achieving approximately 20x faster inference than SSA in the closed-set setting. We also show that SFS effectively handles open-vocabulary segmentation by leveraging CLIP-based semantic heads, outperforming recent open-vocabulary models on broad class labeling. This work enables practical real-time semantic segmentation with the "segment-anything" capability, broadening the applicability of foundation segmentation models in robotics scenarios. The implementation is available at https://github.com/KBH00/Semantic-Fast-SAM.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Semantic Segmentation on Cityscapes (val) (mIoU only)](<https://www.sota2.com/research/sota/semantic-segmentation-on-cityscapes-val-miou-only.md>): Semantic segmentation; Cityscapes (val); mIoU: 70.33.

- [Semantic Segmentation on ADE20K (val)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-val.md>): Semantic segmentation; ADE20K (val); mIoU: 48.01.

- [Semantic Segmentation on Pascal VOC (Seen/Unseen Evaluation)](<https://www.sota2.com/research/sota/semantic-segmentation-on-pascal-voc-seen-unseen-evaluation.md>): Semantic segmentation; Pascal VOC; Seen mIoU: not recorded.

- [Semantic Segmentation on ADE20K (Novel and Zero-Shot mIoU)](<https://www.sota2.com/research/sota/semantic-segmentation-on-ade20k-novel-and-zero-shot-miou.md>): Semantic segmentation; ADE20K; mIoU (novel): not recorded.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2604.20169>)

[PDF](<https://arxiv.org/pdf/2604.20169.pdf>)