ViewCrafter: Taming Video Diffusion Models for High-fidelity Novel View Synthesis

TL;DR

ViewCrafter introduces a novel approach combining video diffusion models and point-based 3D clues to synthesize high-fidelity, consistent novel views from sparse images, enabling immersive and scene-level text-to-3D applications.

Contribution

The paper presents a new method that leverages video diffusion models with iterative view synthesis and camera planning to improve novel view synthesis from limited input images.

Findings

Demonstrates strong generalization across diverse datasets.

Achieves high-fidelity and consistent novel view synthesis.

Enables real-time rendering and scene-level text-to-3D generation.

Abstract

Despite recent advancements in neural 3D reconstruction, the dependence on dense multi-view captures restricts their broader applicability. In this work, we propose \textbf{ViewCrafter}, a novel method for synthesizing high-fidelity novel views of generic scenes from single or sparse images with the prior of video diffusion model. Our method takes advantage of the powerful generation capabilities of video diffusion model and the coarse 3D clues offered by point-based representation to generate high-quality video frames with precise camera pose control. To further enlarge the generation range of novel views, we tailored an iterative view synthesis strategy together with a camera trajectory planning algorithm to progressively extend the 3D clues and the areas covered by the novel views. With ViewCrafter, we can facilitate various applications, such as immersive experiences with real-time rendering by efficiently optimizing a 3D-GS representation using the reconstructed 3D points and the generated novel views, and scene-level text-to-3D generation for more imaginative content creation. Extensive experiments on diverse datasets demonstrate the strong generalization capability and superior performance of our method in synthesizing high-fidelity and consistent novel views.