Abstract

This paper presents an approach for reconstruction of4D temporally coherent models of complex dynamic scenes.No prior knowledge is required of scene structure or cameracalibration allowing reconstruction from multiple movingcameras. Sparse-to-dense temporal correspondence is inte-grated with joint multi-view segmentation and reconstruc-tion to obtain a complete 4D representation of static anddynamic objects. Temporal coherence is exploited to over-come visual ambiguities resulting in improved reconstruc-tion of complex scenes. Robust joint segmentation and re-construction of dynamic objects is achieved by introducinga geodesic star convexity constraint. Comparative evaluation is performed on a variety of unstructured indoor and outdoor dynamic scenes with hand-held cameras and multiple people. This demonstrates reconstruction of complete temporally coherent 4D scene models with improved nonrigid object segmentation and shape reconstruction.