论文

MAVE:从电影镜头自动重建可编辑 3D 布局场景

MAVE: Reconstructing Editable 3D Blockouts from Film Shots

精选理由

用 DeepSeek-1017 当 Agent,每个镜头 0.054 美元就能把视频转成可改的 Blender 场景,时间省一半。做预演或想给视频生成模型喂结构条件的人可以试试,代码和 3K 数据集都要开源。

论文提出 MAVE 框架,可从电影/视频镜头自动重建可编辑的 3D blockout 场景,并配套发布 MAVEBench 基准,包含 50 个电影镜头和 416 个关键帧,用于评估相机运动、空间布局和人物走位。框架分三步:先在共享度量坐标系中恢复相机、人体运动与几何证据;再由 Agent 依据这些约束构建可编辑环境;最后通过程序化检查报告不一致之处并迭代修正,输出可编辑的 Blender 场景与帧对齐的 blockout 视频。

正文

在 MAVEBench 上,MAVE 让三个 Agent 相比各自 vanilla 基线提升 27.5-41.1 分。搭配 DeepSeek-1017 时取得 73.2 分,平均每个镜头 API 成本仅 0.054 美元,总重建时间减少 53%。团队还构建了包含 3,000 对 blockout-视频的 MAVEData-3K 数据集用于模型训练,并将公开发布。创作者可修改视点与角色路径后渲染,用于条件化视频生成,代码已在 GitHub 开源。

原文 · arXiv: DeepSeek

MAVE: Reconstructing Editable 3D Blockouts from Film Shots

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.

产业坐标与信源核验

本条情报归属 AITOP「论文」核心栏目,重点聚焦中国 AI 产业与国产大模型生态。在跨区域产业观察坐标中,该动态直接关联「AI 开发者工具链、Coding Agent 与工程化实践」研究主线,反映出当前技术演进与商业落地的最新风向。涉及核心实体与技术议题包括:MAVE、DeepSeek、Blender、视频生成,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。

信源核验说明:本文原始信息发布自一手权威信源「arXiv: DeepSeek」(Tier-1 一手发布),具备高确定性参考价值。