TrajectoryCrafter: Redirecting Camera Trajectory for Monocular Videos via Diffusion Models
This addresses the challenge of camera trajectory control in monocular videos for applications like video editing and virtual reality, representing a novel method rather than an incremental improvement.
The paper tackles the problem of redirecting camera trajectories in monocular videos by introducing TrajectoryCrafter, which uses a dual-stream conditional video diffusion model to achieve precise control and coherent 4D content generation, demonstrating superior performance in evaluations.
We present TrajectoryCrafter, a novel approach to redirect camera trajectories for monocular videos. By disentangling deterministic view transformations from stochastic content generation, our method achieves precise control over user-specified camera trajectories. We propose a novel dual-stream conditional video diffusion model that concurrently integrates point cloud renders and source videos as conditions, ensuring accurate view transformations and coherent 4D content generation. Instead of leveraging scarce multi-view videos, we curate a hybrid training dataset combining web-scale monocular videos with static multi-view datasets, by our innovative double-reprojection strategy, significantly fostering robust generalization across diverse scenes. Extensive evaluations on multi-view and large-scale monocular videos demonstrate the superior performance of our method.