Research map / Detection and segmentation / Open-set and few-shot detection
On the Transfer of Object-Centric Representation Learning
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Studies how object-centric representation learning transfers to out-of-distribution scenes, building on DINO-feature-based object-centric methods.
Problem
Object-centric methods have mostly been evaluated in-distribution, so their ability to transfer to new data distributions is unclear.
Approach
Examines transfer of object-centric representations learned from pre-trained foundation model features (such as DINO) to out-of-distribution scenes; details are not available in the abstract provided.
Experimental setup
Builds on: DINO
Closest papers by meaning
- Object-Centric Pretraining via Target Encoder Bootstrapping (ICLR 2025 · similarity 0.91)
- A Data-Centric Revisit of Pre-Trained Vision Models for Robot Learning (CVPR 2025 · similarity 0.89)
- Temporally Consistent Object-Centric Learning by Contrasting Slots (CVPR 2025 · similarity 0.89)
- CTRL-O: Language-Controllable Object-Centric Visual Representation Learning (CVPR 2025 · similarity 0.87)
- Learning to Compose: Improving Object Centric Learning by Injecting Compositionality (ICLR 2024 · similarity 0.87)
- Multi-Part Object Representations via Graph Structures and Co-Part Discovery (NeurIPS 2026 · 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.