MyGo: Consistent and Controllable Multi-View Driving Video Generation with Camera Control

TL;DR

MyGo is an innovative framework for generating multi-view driving videos with enhanced camera control and spatial-temporal consistency, advancing autonomous driving simulation capabilities.

Contribution

It introduces camera motion conditions into a pre-trained diffusion model and employs epipolar constraints for improved multi-view consistency.

Findings

Achieved state-of-the-art results in camera-controlled video generation.

Enhanced multi-view driving video generation quality.

Demonstrated improved spatial-temporal consistency in generated videos.

Abstract

High-quality driving video generation is crucial for providing training data for autonomous driving models. However, current generative models rarely focus on enhancing camera motion control under multi-view tasks, which is essential for driving video generation. Therefore, we propose MyGo, an end-to-end framework for video generation, introducing motion of onboard cameras as conditions to make progress in camera controllability and multi-view consistency. MyGo employs additional plug-in modules to inject camera parameters into the pre-trained video diffusion model, which retains the extensive knowledge of the pre-trained model as much as possible. Furthermore, we use epipolar constraints and neighbor view information during the generation process of each view to enhance spatial-temporal consistency. Experimental results show that MyGo has achieved state-of-the-art results in both general camera-controlled video generation and multi-view driving video generation tasks, which lays the foundation for more accurate environment simulation in autonomous driving. Project page: https://metadrivescape.github.io/papers_project/MyGo/page.html