4D Visualization of Dynamic Events from Unconstrained Multi-View Videos
This provides a method for immersive visualization and editing of dynamic events from casual video captures, which is incremental as it builds on existing multi-view and neural rendering techniques.
The paper tackles the problem of creating 4D visualizations from unconstrained multi-view videos by using self-supervised neural networks to model static and dynamic aspects, enabling continuous exploration of space-time and editing features like revealing occluded objects, validated on events captured with up to 15 mobile cameras.
We present a data-driven approach for 4D space-time visualization of dynamic events from videos captured by hand-held multiple cameras. Key to our approach is the use of self-supervised neural networks specific to the scene to compose static and dynamic aspects of an event. Though captured from discrete viewpoints, this model enables us to move around the space-time of the event continuously. This model allows us to create virtual cameras that facilitate: (1) freezing the time and exploring views; (2) freezing a view and moving through time; and (3) simultaneously changing both time and view. We can also edit the videos and reveal occluded objects for a given view if it is visible in any of the other views. We validate our approach on challenging in-the-wild events captured using up to 15 mobile cameras.