Abstract
Existing dynamic scene generation methods mostly rely on distilling knowledge from pre-trained 3D generative models, which are typically fine-tuned on synthetic object datasets. As a result, the generated scenes are often object-centric and lack photorealism. To address these limitations, we introduce a novel pipeline designed for photorealistic text-to-4D scene generation, discarding the dependency on multi-view generative models and instead fully utilizing video generative models trained on diverse real-world datasets. Our method begins by generating a reference video using the video generation model. We then learn the canonical 3D representation of the video using a freeze-time video, delicately generated from the reference video. To handle inconsistencies in the freeze-time video, we jointly learn a per-frame deformation to model these imperfections. We then learn the temporal deformation based on the canonical representation to capture dynamic interactions in the reference video. The pipeline facilitates the generation of dynamic scenes with enhanced photorealism and structural integrity, viewable from multiple perspectives, thereby setting a new standard in 4D scene generation.
Abstract (translated)
现有的动态场景生成方法主要依赖于从预训练的3D生成模型中提取知识，这些模型通常在合成物体数据集上进行微调。因此，生成的场景通常是以物体为中心，缺乏真实感。为了克服这些限制，我们引入了一种专为 photorealistic 文本到4D场景生成而设计的新流程，摒弃了多视角生成模型的依赖，而是完全利用基于多样化现实数据集训练的 video 生成模型。 我们的方法首先使用视频生成模型生成参考视频。然后，我们通过一个由参考视频生成的静态视频来学习视频的规范3D表示。为了处理静态视频中的不一致性，我们与合作学习每个帧的变形来建模这些不完美之处。接下来，根据规范表示学习时间变形，以捕捉参考视频中的动态交互。该流程使得可以从多个角度生成具有更高真实感和结构完整性的动态场景，为4D场景生成树立了新的标准。
URL
https://arxiv.org/abs/2406.07472
