Research map / Detection and segmentation / Weakly supervised detection and segmentation
ProMerge: Prompt and Merge for Unsupervised Instance Segmentation
Dylan J Li, Gyungin Shin
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 background-based mask pruning to prompt-and-merge (ablation) | ProMerge · COCO2017 | mask AP: 0.4 -> 2.1 | Improved |
| add connected-component splitting (ablation) | ProMerge · COCO2017 | mask AR: 4.7 -> 7.5 | Improved |
| add merging to prompting without pruning (ablation) | ProMerge · COCO2017 | mask AP: 0.5 -> 0.4 | Hurt |
| add pixel-wise voting for background aggregation (ablation) | ProMerge · COCO2017 | mask AP: 1.2 -> 2.4 | Improved |
| use Cascade filtering instead of IoA or feature-similarity filtering (ablation) | ProMerge · COCO2017 | mask AP: 1.7 / 2.0 -> 2.4 | Improved |
| use both feature similarity and IoA merging conditions (ablation) | ProMerge · COCO2017 | mask AP: 2.2 -> 2.4 | Improved |
| train detector on ProMerge pseudo-labels instead of MaskCut pseudo-labels | Cascade Mask R-CNN · Average over 6 benchmarks | AP: 7.0 -> 7.9 | Improved |
| replace normalized-cut with prompt-and-merge | unsupervised instance segmentation · COCO2017 | speed: 0.54 FPS, about 1.6x TokenCut and 3.6x MaskCut | Improved |
| increase stride beyond 4 to 15 or 30 | ProMerge · COCO2017 | AP/AR | Hurt |
Later papers that compared with this method
3 later accepted papers report ProMerge as a baseline in their results tables; 3 of them report a better number on most of the results they share with it.
- TRACE: Your Diffusion Model is Secretly an Instance Edge Detector (ICLR 2026 · better on 2 of 2 shared results)
- Beyond Single Images: Retrieval Self-Augmented Unsupervised Camouflaged Object Detection (ICCV 2025 · better on 9 of 9 shared results)
- Ensemble Foreground Management for Unsupervised Object Discovery (ICCV 2025 · better on 1 of 1 shared result)
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.
- TokenCut — TokenCut: Segmenting Objects in Images and Videos With Self-Supervised Transformer and Normalized Cut (TPAMI 2023) · better on 1 of 1 shared result
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
- SOLO: A Simple Framework for Instance Segmentation (TPAMI 2021 · similarity 0.90)
- SOLO: Segmenting Objects by Locations (ECCV 2020 · similarity 0.89)
- The Devil is in the Object Boundary: Towards Annotation-free Instance Segmentation using Foundation Models (ICLR 2024 · similarity 0.88)
- Segment Anything without Supervision (NeurIPS 2024 · similarity 0.88)
- Training-Free Dataset Pruning for Instance Segmentation (ICLR 2025 · similarity 0.87)
- Unsupervised Universal Image Segmentation (CVPR 2024 · 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.
