Research map / Detection and segmentation / Open-set and few-shot detection
PARTS: Unsupervised Segmentation With Slots, Attention and Independence Maximization
Daniel Zoran, Rishabh Kabra, Alexander Lerchner, Danilo J. Rezende
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 top-down information (previous action and loss gradient) to slot-attention queries (ablation) | PARTS · PLAYROOM | ARI: 0.72 -> 0.78 | Improved |
| use a fixed independent prior instead of a predictive prior (as in OP3) (ablation) | PARTS · PLAYROOM unrolled 32 steps on static scene | segmentation quality / KL | Improved |
| use a predictive prior instead of a fixed prior (ablation) | PARTS · PLAYROOM unrolled 32 steps on static scene | reconstruction error | Improved |
| build from OP3 up to full PARTS (add slot attention, fixed prior, actions, transformer+LSTM transition) (ablation) | OP3 · PLAYROOM ablation (16-step) | ARI-F: 0.425 -> 0.712 | Improved |
| replace S-IODINE/OP3 with PARTS | S-IODINE · PLAYROOM | ARI-F: 0.4461 -> 0.7349 | Improved |
| use PARTS instead of OAT | OAT · PLAYROOM | ARI-F: 0.6053 -> 0.7349 | Improved |
| use PARTS instead of OAT | OAT · PLAYROOM | reconstruction MSE: 0.0025 -> 0.0023 | Improved |
Closest papers by meaning
- Unsupervised Part Discovery from Contrastive Reconstruction (NeurIPS 2021 · similarity 0.89)
- Unsupervised Learning of 3D Structure from Images (NeurIPS 2016 · similarity 0.89)
- Unsupervised Co-part Segmentation through Assembly (ICML 2021 · similarity 0.89)
- PartCrafter: Structured 3D Mesh Generation via Compositional Latent Diffusion Transformers (NeurIPS 2025 · similarity 0.88)
- PARTONOMY: Large Multimodal Models with Part-Level Visual Understanding (NeurIPS 2025 · similarity 0.88)
- Hierarchy-Agnostic Unsupervised Segmentation: Parsing Semantic Image Structure (NeurIPS 2024 · similarity 0.88)
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.
