Faster Segment Anything: Towards Lightweight SAM for Mobile Applications
Paper • 2306.14289 • Published • 16
Pretrained weights for MobileSAM,
used by kornia.models.SegmentAnything.
MobileSAM replaces the heavy ViT-H image encoder of SAM with a distilled TinyViT-5M encoder, reducing model size by 60× while retaining most of SAM's segmentation quality. arXiv 2023.
Original repo: ChaoningZhang/MobileSAM
| File | Description |
|---|---|
mobile_sam.pt |
TinyViT-5M encoder + SAM decoder |
@article{zhang2023mobilesam,
title = {Faster Segment Anything: Towards Lightweight SAM
for Mobile Applications},
author = {Zhang, Chaoning and Han, Dongshen and Qiao, Yu
and Kim, Jung Uk and Bae, Sung-Ho and Lee, Seungkyu
and Hong, Choong Seon},
journal = {arXiv preprint arXiv:2306.14289},
year = {2023}
}