MultiCamVideo: Synchronized Multi-Camera UE5 Videos and Trajectories

MultiCamVideo is a synthetic dataset of synchronized multi-camera videos and corresponding camera trajectories rendered in Unreal Engine 5. Created by KlingTeam and released in 2025 alongside the ReCamMaster paper, it provides ground-truth spatial data for multi-view video research.

Use Cases

• Training generative models for camera-controlled video generation using synchronized frames
• Developing 3D and 4D reconstruction algorithms using ground-truth camera trajectories
• Benchmarking multi-view consistency in video production pipelines

Strengths

• Rendered in Unreal Engine 5 for high-fidelity visual simulation
• Provides perfectly synchronized multi-view frames
• Includes exact ground-truth camera trajectories

Limitations

• Synthetic nature may lack real-world camera artifacts like motion blur or sensor noise
• Limited to virtual environments which may not capture the full complexity of natural lighting

Provenance

Source

KlingTeam, ReCamMaster (Arxiv 2503.11647)

Collection Method

Synthetic rendering using Unreal Engine 5

Freshness

Last updated October 2025.