Research map / Detection and segmentation / CNN features
LiteSAM is Actually what you Need for segment Everything
Jianhai Fu, Yuanjie Yu, Ningchuan Li, Yi Zhang, Qichao Chen, Jianping Xiong, Jun Yin, Zhiyu Xiang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace grid-search prompt sampling with AutoPPN on SAM-B (ablation) | SAM-B · COCO SegEvery | SegEvery time: 2084 ms -> 120 ms | Improved |
| replace grid-search prompt sampling with AutoPPN on SAM-B (ablation) | SAM-B · COCO SegEvery | AR@1000: 55.1 -> 54.7 | Hurt |
| replace grid search with AutoPPN on LiteViT (ablation) | LiteViT · COCO SegEvery | SegEvery time: 1320 ms -> 80 ms | Improved |
| replace grid search with AutoPPN on MobileSAM (ablation) | MobileSAM · COCO SegEvery | SegEvery time: 2500 ms -> 115 ms | Improved |
| add stem MSPM, new GT and loss, and object grouping to PPN (ablation) | AutoPPN · COCO SegEvery | Mask AR@1000: 48.8 -> 53.0 | Improved |
| add object grouping to MSPM stem (ablation) | AutoPPN · COCO SegEvery | Mask AR@1000: 49.5 -> 52.3 | Improved |
| use Multi-Scale Pooling Self Attention block instead of original PoolFormer block (ablation) | LiteViT · COCO 1-box prompt | mAP: 52.7 -> 55.3 | Improved |
| prune embedding dims of PoolFormer-S12 to 1.15M parameters (ablation) | PoolFormer-S12 · COCO 1-box prompt | mAP: 55.1 -> 52.7 | Hurt |
| scale LiteViT to about 2x and 3x parameters (ablation) | LiteViT · COCO 1-box prompt | mAP: 55.3 -> 56.9 -> 58.1 | Improved |
Closest papers by meaning
- Bridge the Points: Graph-based Few-shot Segment Anything Semantically (NeurIPS 2024 · similarity 0.88)
- TinySAM: Pushing the Envelope for Efficient Segment Anything Model (AAAI 2025 · similarity 0.88)
- Stable Segment Anything Model (ICLR 2025 · similarity 0.87)
- AoP-SAM: Automation of Prompts for Efficient Segmentation (AAAI 2025 · similarity 0.87)
- SAM 2: Segment Anything in Images and Videos (ICLR 2025 · similarity 0.87)
- SAM-CP: Marrying SAM with Composable Prompts for Versatile Segmentation (ICLR 2025 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
