AI Paper Atlas

Research map / Video / Video-language understanding

Natural Language Camera Movement Understanding

Yuwen Tan, Joey Huang, Jin Huang, Haoxiang Li, Boqing Gong

ECCV 2026 · Video › Video-language understanding · approach: Video LLMs with fine-grained understanding · 1 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 targeted re-sampling to balance categories (ablation)SFT Qwen3-VL-4B · ACaM real-worldaverage class accuracy: 48.42 -> 63.78Improved
add targeted re-sampling to balance categories (ablation)SFT Qwen3-VL-4B · ACaM syntheticaverage class accuracy: 58.56 -> 72.07Improved
add data augmentation (zoom crop, reversal, flip, roll) on top of sampling (ablation)SFT Qwen3-VL-4B · ACaM real-worldaverage class accuracy: 63.78 -> 70.88Improved
add data augmentation on top of sampling (ablation)SFT Qwen3-VL-4B · ACaM syntheticaverage class accuracy: 72.07 -> 75.03Improved
raise LoRA rank from 64 to 256 (ablation)SFT Qwen3-VL-4B · ACaM real-worldaverage class accuracy: 70.88 -> 72.27Improved
raise LoRA rank from 64 to 256 (ablation)SFT Qwen3-VL-4B · ACaM syntheticaverage class accuracy: 75.03 -> 74.40No clear effect
raise LoRA rank from 64 to 256 (ablation)SFT Qwen3-VL-8B · ACaM real-worldoverall accuracy: 68.10 -> 72.75Improved
raise LoRA rank from 64 to 256 (ablation)SFT Qwen3-VL-8B · ACaM syntheticoverall accuracy: 70.91 -> 78.20Improved
scale SFT model from 4B to 8BSFT Qwen3-VL · ACaM syntheticoverall accuracy: 73.28 -> 78.20Improved
instruction-tune Qwen3-VL-8B on ACaM training setQwen3-VL-8B · ACaM real-worldoverall accuracy: 58.27 -> 72.75Improved
increase input frames from 8 to 32 for low-intensity push in (ablation)Gemini-3-Pro · Push-in low intensityaccuracy: 29.17 -> 29.17No clear effect
increase camera movement intensity from low to highQwen3-VL-32B · Push-in (8 frames)accuracy: 41.67 -> 100.00Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 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