AI Paper Atlas

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

ECCV 2022 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Embodied agent learning · 5 citations (OpenAlex)

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 changedWhereMeasuredEffect
add separate segmentation and target localization to a policy that learns from images directly (ablation)No Mask Baseline · ManipulaTHOR APND testSR: 0.0 -> 11.6Improved
add agent relative location as input (Loc-MDM vs MDM) (ablation)MDM · ManipulaTHOR APND test (predicted masks)SR: 1.62 -> 3.24Improved
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 -> 11.6Improved
use m-VOLE location estimation instead of Loc-MDM with GT masks (ablation)Loc-MDM · ManipulaTHOR APND test (GT masks)SR: 21.6 -> 59.6Improved
use conditional segmentation instead of MaskRCNN detection (ablation)m-VOLE · ManipulaTHOR APND testSR: 8.7 -> 11.6Improved
add temporal location estimation under depth noise (m-VOLE vs Loc-MDM)Loc-MDM · ManipulaTHOR APND test with depth noiseSR relative drop: -64.8% -> -47.7%Improved
evaluate on novel objects instead of seen objectsm-VOLE · ManipulaTHOR APND testSR: 18.7 -> 5.24Hurt
use m-VOLE instead of Loc-MDM on novel objectsLoc-MDM · ManipulaTHOR APND test, novel objectsSR: 0.9 -> 5.24Improved
add arm control reward (ablation)m-VOLE · ManipulaTHOR APND test (GT masks)SR: 0 -> 59.6Improved
add exploration reward (ablation)m-VOLE · ManipulaTHOR APND test (GT masks)SR: 15.4 -> 59.6Improved
add object visibility reward (ablation)m-VOLE · ManipulaTHOR APND test (GT masks)SR: 53.6 -> 59.6Improved
use m-VOLE instead of ArmPointNav with perfect GPS and compassArmPointNav · ManipulaTHOR APND test (GT masks)SR: 58.3 -> 59.6Improved

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy