AI Paper Atlas

Research map / Video / Video-language understanding

HanDyVQA: A Video QA Benchmark for Fine-Grained Hand-Object Interaction Dynamics

Masatoshi Tateno, Gido Kato, Hirokatsu Kataoka, Yoichi Sato, Takuma Yagi

CVPR 2026 · Video › Video-language understanding · approach: Event and anomaly reasoning in video · 4 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
fine-tune InternVideo2 on the 1.1K training questions (ablation)InternVideo2-Stage2 · HanDyVQA MCQAvg accuracy: 33.3 -> 51.0Improved
add 3D hand pose input to fine-tuned InternVideo2 (ablation)InternVideo2-Stage2 fine-tuned · HanDyVQA MCQAvg accuracy: 51.0 -> 53.3Improved
add object features to fine-tuned InternVideo2 (ablation)InternVideo2-Stage2 fine-tuned · HanDyVQA MCQAvg accuracy: 51.0 -> 53.3Improved
add object bbox tracklets to fine-tuned InternVideo2 (ablation)InternVideo2-Stage2 fine-tuned · HanDyVQA MCQAvg accuracy: 51.0 -> 52.4Improved
combine all HOI cues (hand + bbox + features) instead of fewer cues (ablation)InternVideo2-Stage2 fine-tuned · HanDyVQA MCQAvg accuracy: Hand+ObjFeats 53.9 vs all three 53.4No clear effect
increase input frames and resolution (ablation)Qwen2.5-VL-7B, Gemini-2.5-Pro · HanDyVQA MCQaccuracyImproved
increase input frames beyond 8 (ablation)mPLUG-Owl3-8B · HanDyVQA MCQaccuracyNo clear effect
increase input frames and resolution on motion errors (ablation)MLLMs · HanDyVQA MCQMotion error rateNo clear effect
use hand/object-aware HelpingHands instead of LaViLaLaViLa-L (4 frames) · HanDyVQA MCQ zero-shotAvg accuracy: 41.8 -> 40.3 avg; Objects 67.5 -> 69.2Mixed
use hand/object-aware EgoHOD instead of LaViLaLaViLa-L (16 frames) · HanDyVQA MCQ zero-shotAvg accuracy: Objects 68.5 -> 75.3; Avg 42.6 -> 42.7Mixed
use video input instead of frame-wise image input for ReasoningVOS (ablation)Sa2VA-8B · HanDyVQA ReasoningVOS ObjectsJ: Objects J all 36.2 (image) vs 32.0 (video)Mixed
scale Sa2VA from 4B to 8B (ablation)Sa2VA · HanDyVQA ReasoningVOS ObjectsJ (All): 17.2 -> 32.0Improved

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