cs.CVNov 17, 2026

MAVE: Reconstructing Editable 3D Blockouts from Film Shots

Authors: Yifan Chang, Jian Yang, Jiangnan Shao, Xiaofeng Wang, Zhaoxin Fan, Renjing Xu, Ming Lu, Xingang Wang, 

Organizations: GigaAI · CASIA · UCAS · RUC · 

Abstract

Animated 3D blockouts support previsualization and provide structural conditions for video generation, but manually reconstructing them from reference videos is costly. This limits both their use by creators and the production of paired training data. We introduce MAVEBench, a benchmark of 50 film shots and 416 keyframes for evaluating camera motion, spatial layout, and character blocking. We propose MAVE, a video-to-blockout framework combining Metric observation, Agentic modeling, and evidence-based VErification. It first recovers cameras, human motion, and geometric evidence in a shared metric coordinate system. An agent then constructs an editable environment under constraints from these observations, while programmatic checks report inconsistencies to guide refinement. The outputs are editable Blender scenes and frame-aligned blockout videos. On MAVEBench, MAVE improves overall scores by 27.5-41.1 points across three agents over their corresponding vanilla baselines. With DeepSeek-1017, it achieves a score of 73.2 at an average API cost of only USD 0.054 per shot, outperforming the strongest vanilla baseline while reducing total reconstruction time by 53%. We also construct MAVEData-3K, a dataset of 3,000 blockout-video pairs for model training, and will release it publicly. Creators can change viewpoints and character paths, then render the edited scenes to condition video generation. Code is available at https://github.com/yfChang-cv/MAVE.

Figures & tables

Explore similar work

CardsList
  1. CineRecon: Benchmarking Frontier Models for Agentic Cinematic Reconstruction

    Oct 12, 2026Jialei Wang, Zehan Wang, XiaoYang Wu +63D Reconstruction BenchmarksDynamic Scene Reconstruction

  2. SceneAE: Scene Generation and Reconstruction in Queryable 3D-Anchored Latent Space

    Oct 14, 2026Shuofeng Sun, Wenzhao Zheng, Yunpeng Zhang +33D Scene Generation3D Scene Representation

  3. SceneAnything: Agentic Reconstruction of Large-Scale Mesh Scenes from a Single Video

    Nov 19, 2026Bowen Xue, Saeed Hadadan, Fabrice Rousselle +33D Reconstruction3D Mesh Reconstruction