Research map / Detection and segmentation / Open-set and few-shot detection
AIMS: All-Inclusive Multi-Level Segmentation for Anything
Lu Qi, Jason Kuen, Weidong Guo, Jiuxiang Gu, Zhe Lin, Bo Du, Yu Xu, Ming-Hsuan Yang
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 |
|---|---|---|---|
| train on all five datasets instead of PPP only | baseline decoders · COCO-PSG | AP_entity: 25.7 -> 40.4 | Improved |
| train on all datasets instead of COCO-PSG only | baseline decoders · COCO-PSG | AP_relation: 39.6 -> 38.9 | Hurt |
| add mask prompt encoder (ablation) | baseline · PPP | AP_part: 24.5 -> 25.7 | Improved |
| add task complementarity module (ablation) | baseline · PPP | AP_part: 24.5 -> 25.9 | Improved |
| add association module (ablation) | baseline · PPP | AR_EP@100: 69.7 -> 70.6 | Improved |
| use only full-image mask prompts in training (ablation) | AIMS · PPP | AP_entity: 53.2 -> 53.4 | No clear effect |
| use only partial-image mask prompts in training (ablation) | AIMS · PPP | AP_part: 24.5 -> 22.1 | Hurt |
| use both full and partial image prompts (ablation) | AIMS · PPP | AP_part: 24.5 -> 25.6 | Improved |
| add single-entity and dual-entity mask prompts to full and partial prompts (ablation) | AIMS · PPP | AP_part: 25.6 -> 26.5 | Improved |
| use only dual/single-entity prompts without full or partial prompts (ablation) | AIMS · PPP | AP_part: 24.5 -> 20.4 | Hurt |
| add a second CONV1 layer to mask encoder (ablation) | mask prompt encoder · PPP | AP_part: 26.5 -> 26.4 | No clear effect |
| remove CONV1 from mask encoder (ablation) | mask prompt encoder · PPP | AP_part: 26.5 -> 25.9 | 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.
- PanopticPartFormer++ — Panoptic-PartFormer: Learning a Unified Model for Panoptic Part Segmentation (ECCV 2022) · better on 3 of 3 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
- Segment Anyword: Mask Prompt Inversion for Open-Set Grounded Segmentation (ICML 2025 · similarity 0.86)
- Bridge the Points: Graph-based Few-shot Segment Anything Semantically (NeurIPS 2024 · similarity 0.86)
- Semantic-aware SAM for Point-Prompted Instance Segmentation (CVPR 2024 · similarity 0.85)
- Segment Anything without Supervision (NeurIPS 2024 · similarity 0.85)
- Reverse Region-to-Entity Annotation for Pixel-Level Visual Entity Linking (AAAI 2025 · similarity 0.85)
- USE: Universal Segment Embeddings for Open-Vocabulary Image Segmentation (CVPR 2024 · similarity 0.85)
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.
