Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Object Manipulation via Visual Target Localization
Kiana Ehsani, Ali Farhadi, Aniruddha Kembhavi, Roozbeh Mottaghi
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 separate segmentation and target localization to a policy that learns from images directly (ablation) | No Mask Baseline · ManipulaTHOR APND test | SR: 0.0 -> 11.6 | Improved |
| add agent relative location as input (Loc-MDM vs MDM) (ablation) | MDM · ManipulaTHOR APND test (predicted masks) | SR: 1.62 -> 3.24 | Improved |
| use relative location to aggregate target location over time (m-VOLE) instead of feeding it naively (Loc-MDM) (ablation) | Loc-MDM · ManipulaTHOR APND test (predicted masks) | SR: 3.24 -> 11.6 | Improved |
| use m-VOLE location estimation instead of Loc-MDM with GT masks (ablation) | Loc-MDM · ManipulaTHOR APND test (GT masks) | SR: 21.6 -> 59.6 | Improved |
| use conditional segmentation instead of MaskRCNN detection (ablation) | m-VOLE · ManipulaTHOR APND test | SR: 8.7 -> 11.6 | Improved |
| add temporal location estimation under depth noise (m-VOLE vs Loc-MDM) | Loc-MDM · ManipulaTHOR APND test with depth noise | SR relative drop: -64.8% -> -47.7% | Improved |
| evaluate on novel objects instead of seen objects | m-VOLE · ManipulaTHOR APND test | SR: 18.7 -> 5.24 | Hurt |
| use m-VOLE instead of Loc-MDM on novel objects | Loc-MDM · ManipulaTHOR APND test, novel objects | SR: 0.9 -> 5.24 | Improved |
| add arm control reward (ablation) | m-VOLE · ManipulaTHOR APND test (GT masks) | SR: 0 -> 59.6 | Improved |
| add exploration reward (ablation) | m-VOLE · ManipulaTHOR APND test (GT masks) | SR: 15.4 -> 59.6 | Improved |
| add object visibility reward (ablation) | m-VOLE · ManipulaTHOR APND test (GT masks) | SR: 53.6 -> 59.6 | Improved |
| use m-VOLE instead of ArmPointNav with perfect GPS and compass | ArmPointNav · ManipulaTHOR APND test (GT masks) | SR: 58.3 -> 59.6 | Improved |
Closest papers by meaning
- AffordDP: Generalizable Diffusion Policy with Transferable Affordance (CVPR 2025 · similarity 0.86)
- EC-Diffuser: Multi-Object Manipulation via Entity-Centric Behavior Generation (ICLR 2025 · similarity 0.85)
- BOP-ASK: Object-Interaction Reasoning for Vision-Language Models (CVPR 2026 · similarity 0.85)
- Entity-Centric Reinforcement Learning for Object Manipulation from Pixels (ICLR 2024 · similarity 0.85)
- Learning Spatial-Aware Manipulation Ordering (NeurIPS 2025 · similarity 0.85)
- OSVI-WM: One-Shot Visual Imitation for Unseen Tasks using World-Model-Guided Trajectory Generation (NeurIPS 2025 · similarity 0.85)
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.
