AI Paper Atlas

Research map / 3D vision / 3D generation

Embodied VideoAgent: Persistent Memory from Egocentric Videos and Embodied Sensors Enables Dynamic Scene Understanding

Yue Fan, Xiaojian Ma, Rongpeng Su, Jun Guo, Rujie Wu, Xi Chen, Qing Li

ICCV 2025 · 3D vision › 3D generation · approach: Scene-aware human motion synthesis · 21 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
use visual similarity for object re-ID instead of text only (ablation)Embodied VideoAgent · Ego4D VQ3D valSucc%: 53.05 -> 85.37Improved
use visual similarity for object re-ID instead of text only (ablation)Embodied VideoAgent · Ego4D VQ3D valQwP%: 56.10 -> 92.07Improved
use visual similarity for object re-ID instead of text only (ablation)Embodied VideoAgent · Ego4D VQ3D valSucc*%: 94.57 -> 92.72Hurt
use open-vocabulary detector with aggressive predictions vs EgoLocEmbodied VideoAgent(image) · Ego4D VQ3D valSucc%: 80.49 -> 85.37Improved
use open-vocabulary detector vs EgoLocEmbodied VideoAgent(image) · Ego4D VQ3D valL2: 1.45 -> 1.86Hurt
swap InternVL2-8B vqa tool for GPT-4o (ablation)Embodied VideoAgent · OpenEQA subsetaccuracy ALL: 41.2 -> 47.0Improved
use persistent object memory agent instead of VideoAgentEmbodied VideoAgent(InternVL2-8B) · OpenEQA subsetaccuracy ALL: 36.3 -> 41.2Improved
use agentic system instead of end-to-end VLMEmbodied VideoAgent(GPT-4o) · OpenEQA subsetaccuracy ALL: 30.6 -> 47.0Improved
replace GPT-4 w/ CG scene graph with frame captions (LLaVA-1.5)GPT-4 · OpenEQA fullaccuracy ALL: 36.5 -> 43.6Improved
use Embodied VideoAgent instead of VideoAgentEmbodied VideoAgent · EnvQAEvents accuracy: 5.54 -> 25.91Improved
use Embodied VideoAgent instead of VideoAgentEmbodied VideoAgent · EnvQAStates accuracy: 12.5 -> 35.50Improved
use Embodied VideoAgent instead of VideoAgentEmbodied VideoAgent · EnvQAOrders accuracy: 65.5 -> 68.00Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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