Cavia: Camera-controllable Multi-view Video Diffusion with View-Integrated Attention

TL;DR

Cavia is a novel framework that enables camera-controllable, multi-view video generation with high spatial and temporal consistency, allowing precise camera motion specification alongside object motion, surpassing previous methods in quality.

Contribution

Introduces Cavia, the first framework for joint camera and object motion control in multi-view video generation with view-integrated attention modules.

Findings

Outperforms state-of-the-art in geometric consistency

Achieves high perceptual quality in generated videos

Supports diverse data sources for training

Abstract

In recent years there have been remarkable breakthroughs in image-to-video generation. However, the 3D consistency and camera controllability of generated frames have remained unsolved. Recent studies have attempted to incorporate camera control into the generation process, but their results are often limited to simple trajectories or lack the ability to generate consistent videos from multiple distinct camera paths for the same scene. To address these limitations, we introduce Cavia, a novel framework for camera-controllable, multi-view video generation, capable of converting an input image into multiple spatiotemporally consistent videos. Our framework extends the spatial and temporal attention modules into view-integrated attention modules, improving both viewpoint and temporal consistency. This flexible design allows for joint training with diverse curated data sources, including scene-level static videos, object-level synthetic multi-view dynamic videos, and real-world monocular dynamic videos. To our best knowledge, Cavia is the first of its kind that allows the user to precisely specify camera motion while obtaining object motion. Extensive experiments demonstrate that Cavia surpasses state-of-the-art methods in terms of geometric consistency and perceptual quality. Project Page: https://ir1d.github.io/Cavia/