AI Paper Atlas

Research map / Video / Video-language understanding

Memory Tree Guided Key Frame Querying for Efficient 3D Question Answering

Hsiang-Wei Huang, Fu-Chen Chen, Li-Wu Tsao, Cheng-Han Lee, Che-Chun Su, Lu Xia, Ronghui Peng, Jenq-Neng Hwang, Min Sun, Cheng-Hao Kuo

ECCV 2026 · Video › Video-language understanding · approach: Video question answering · 0 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 MemTree3D key frame selection instead of uniform samplingGPT-4o · OpenEQALLM-Match: 49.4 -> 66.8Improved
use MemTree3D key frame selection instead of uniform samplingLLaVA-OneVision-7B · OpenEQALLM-Match: 49.2 -> 55.0Improved
use MemTree3D instead of uniform samplingLLaVA-OneVision-7B · ScanQAEM@1: 25.1 -> 28.0Improved
use MemTree3D instead of uniform samplingLLaVA-OneVision-7B · SQA3DEM@1: 46.2 -> 49.6Improved
add MemTree3D on larger HM3D scenes versus smaller ScanNet scenesGPT-4o · OpenEQA HM3D vs ScanNetLLM-Match: +24.6 vs +13.6Improved
use 6-DoF pose segmentation instead of uniform 90-frame segmentation (ablation)MemTree3D LocNode · OpenEQALLM-Match: 33.0 -> 49.1Improved
use 6-DoF pose segmentation instead of uniform 30-frame segmentation (ablation)MemTree3D LocNode · OpenEQALLM-Match: 40.4 -> 49.1Improved
shorten uniform segmentation window from 90 to 30 frames (ablation)MemTree3D LocNode · OpenEQALLM-Match: 33.0 -> 40.4Improved
use GPT-4o instead of Qwen3-4B as LLM (ablation)MemTree3D with LLaVA-OneVision-7B · OpenEQALLM-Match: 51.1 -> 55.0Improved
use Qwen3-8B instead of Qwen3-4B as LLM (ablation)MemTree3D with LLaVA-OneVision-7B · OpenEQALLM-Match: 51.1 -> 53.1Improved
use GPT-4o VLM instead of LLaVA-OneVision-7B (ablation)MemTree3D with GPT-4o LLM · OpenEQALLM-Match: 55.0 -> 66.8Improved
query objects absent from scene representation (perception failure)MemTree3D · OpenEQALLM-Match: 69.0 -> 65.9Hurt

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