Research map / Detection and segmentation / Open-set and few-shot detection
Interactron: Embodied Adaptive Object Detection
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Adapt DETR at test time with Interactron's learned loss and learned interaction policy | DETR · AI2-iTHOR test | AP: 0.256 -> 0.328 | Helped |
| Adapt DETR at test time with Interactron's learned loss and learned interaction policy | DETR · AI2-iTHOR test | AP50: 0.448 -> 0.575 | Helped |
| Add test-time adaptive learning using a random exploration policy | DETR · AI2-iTHOR test | AP: 0.288 -> 0.313 | Helped |
| Use a learned policy instead of a random policy to select frames (ablation) | Interactron-Rand · AI2-iTHOR test | AP: 0.313 -> 0.328 | Helped |
| Use a learned policy instead of a random policy to select frames (ablation) | Interactron-Rand · Habitat test (trained on AI2-iTHOR) | AP50: 0.26 -> 0.27 | Helped |
| Add adaptive gradient updates during inference (ablation) | Interactron · AI2-iTHOR test | AP50: 0.28 -> 0.58 | Helped |
| Increase the rollout from 5 to 7 or 9 frames (ablation) | Interactron · AI2-iTHOR test | AP50: 0.575 -> 0.568; 0.575 -> 0.579 | No clear effect |
| Increase the rollout from 5 to 7 or 9 frames (ablation) | Multi-frame · AI2-iTHOR test | AP50: 0.517 -> 0.518; 0.517 -> 0.521 | No clear effect |
| Use distinct frames from the rollout instead of repeated copies of the initial frame (ablation) | Interactron-Rand · AI2-iTHOR test | AP50: 0.51 -> 0.55 | Helped |
| Apply confidence-weighted pseudo-label updates to Multi-frame (ablation) | Multi-frame · AI2-iTHOR test | AP50: +2 points; 0.538 AP50 | Helped |
Closest papers by meaning
- Embodied-DETR: End-to-End Temporal 3D Object Detection in Egocentric Views (ICML 2026 · similarity 0.87)
- Learn How to See: Collaborative Embodied Learning for Object Detection and Camera Adjusting (AAAI 2024 · similarity 0.87)
- LINK: Learning Instance-level Knowledge from Vision-Language Models for Human-Object Interaction Detection (ICLR 2026 · similarity 0.87)
- Test-Time Adaptive Object Detection with Foundation Model (NeurIPS 2025 · similarity 0.85)
- Active Object Detection with Knowledge Aggregation and Distillation from Large Models (CVPR 2024 · similarity 0.85)
- Visual Diversity and Region-aware Prompt Learning for Zero-shot HOI Detection (NeurIPS 2025 · similarity 0.84)
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.