Research map / Detection and segmentation / Open-set and few-shot detection
Unsupervised Segmentation in Real-World Images via Spelke Object Inference
Honglin Chen, Rahul Venkatesh, Yoni Friedman, Jiajun Wu, Joshua B. Tenenbaum, Daniel L. K. Yamins, Daniel M. Bear
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 excitatory messages to KProp (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.036 -> 0.730 | Improved |
| add inhibitory messages to KProp (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.036 -> 0.730 | Improved |
| use l2 normalization instead of softmax in KProp (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.036 -> 0.730 | Improved |
| use Competition readout instead of argmax (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.676 -> 0.730 | Improved |
| add long-range affinity pairs to local pairs (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.700 -> 0.730 | Improved |
| use all affinity pairs instead of local plus sampled long-range (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.730 -> 0.732 | No clear effect |
| increase KProp iterations from 10 to 40 (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.389 -> 0.730 | Improved |
| increase Competition iterations from 1 to 3 (ablation) | EISEN · TDW-Playroom val (motion supervision) | mIoU: 0.729 -> 0.730 | No clear effect |
| add bootstrapping rounds (round 1 to round 3) | EISEN · Bridge | mIoU: 0.336 -> 0.551 | Improved |
| add bootstrapping rounds (round 1 to round 3) | Mask-RCNN · Bridge | mIoU: 0.053 -> 0.102 | Improved |
| add bootstrapping rounds (round 1 to round 3) | Panoptic DeepLab · Bridge | mIoU: 0.051 -> 0.057 | No clear effect |
| replace full supervision with motion-only supervision | Panoptic DeepLab · TDW-Playroom val | mIoU: 0.870 -> 0.620 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- Panoptic DeepLab — Panoptic-DeepLab: A Simple, Strong, and Fast Baseline for Bottom-Up Panoptic Segmentation (CVPR 2020) · better on 6 of 7 shared results
- SSAP — SSAP: Single-Shot Instance Segmentation With Affinity Pyramid (ICCV 2019) · better on 3 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
- Unsupervised Multi-Object Segmentation by Predicting Probable Motion Patterns (NeurIPS 2022 · similarity 0.88)
- Self-Supervised Learning of Object Parts for Semantic Segmentation (CVPR 2022 · similarity 0.87)
- Learning Segmentation from Point Trajectories (NeurIPS 2024 · similarity 0.86)
- Bootstrapping Objectness From Videos by Relaxed Common Fate and Visual Grouping (CVPR 2023 · similarity 0.86)
- Unsupervised Co-part Segmentation through Assembly (ICML 2021 · similarity 0.86)
- GrabS: Generative Embodied Agent for 3D Object Segmentation without Scene Supervision (ICLR 2025 · 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.
