Research map / Detection and segmentation / Open-set and few-shot detection
IP-SAM: Rethinking Prompt-Conditioned Segmentation for Prompt-Absent Deployment
Huiyao Zhang, Jin Bai, Rui Guo, Jianwen Tan, Hongfei Wang, Ye Li
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 |
|---|---|---|---|
| add Self-Prompt Generator (SPG) with prompt-space conditioning (ablation) | IP-SAM baseline · COD10K | MAE: 0.024 -> 0.022 | Improved |
| add Prompt-Space Gating (PSG) after SPG (ablation) | IP-SAM · CAMO | MAE: 0.041 -> 0.034 | Improved |
| add refinement head to full model without it (ablation) | IP-SAM · COD10K | MAE: 0.018 -> 0.017 | Improved |
| route SPG output through frozen prompt encoder and decoder instead of using it directly as mask (ablation) | IP-SAM · COD10K | MAE: 0.025 -> 0.017 | Improved |
| prompt-space conditioning with low-capacity Minimal-SPG (ablation) | Minimal-SPG · COD10K | MAE: 0.051 -> 0.040 | Improved |
| condition through prompt space instead of adding dense prompts to image features (FeatAdd) (ablation) | IP-SAM · COD10K | MAE: 0.020 -> 0.017 | Improved |
| keep prompt encoder frozen instead of trainable (ablation) | IP-SAM · COD10K | MAE: 0.019 -> 0.017 | Improved |
| use asymmetric gating instead of subtraction fusion of Z+ and Z- (ablation) | PSG · COD10K | MAE: 0.019 -> 0.017 | Improved |
| anchor PSG residual to original Z+ instead of suppressed Z+ (ablation) | PSG · COD10K | MAE: 0.019 -> 0.017 | Improved |
| use propagated SPG output tokens instead of default SAM2 mask tokens (ablation) | IP-SAM · COD10K | MAE: 0.019 -> 0.017 | Improved |
| apply LoRA to all blocks instead of deep stages only (ablation) | IP-SAM · COD10K | MAE: 0.034 -> 0.017 | Improved |
| scale backbone down from SAM2-L to SAM2-T (ablation) | IP-SAM · COD10K | MAE: 0.017 -> 0.080 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- ZoomNeXt — ZoomNeXt: A Unified Collaborative Pyramid Network for Camouflaged Object Detection (TPAMI 2024) · better on 4 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Relax Image-Specific Prompt Requirement in SAM: A Single Generic Prompt for Segmenting Camouflaged Objects (AAAI 2024 · similarity 0.87)
- AoP-SAM: Automation of Prompts for Efficient Segmentation (AAAI 2025 · similarity 0.87)
- Stable Segment Anything Model (ICLR 2025 · similarity 0.87)
- Prompting Diffusion Models for Zero-Shot Instance Segmentation (NeurIPS 2026 · similarity 0.86)
- Visual Prompting for Generalized Few-shot Segmentation: A Multi-scale Approach (CVPR 2024 · similarity 0.86)
- UGround: Towards Unified Visual Grounding with Unrolled Transformers (ICML 2026 · similarity 0.86)
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.
