Video inpainting of complex scenes based on local statistical model

This paper describes a framework for temporally consistent video completion. Proposed method allow to remove dynamic objects or restore missing or tainted regions present in a video sequence by utilizing spatial and temporal information from neighboring scenes. The algorithm iteratively performs following operations: achieve frame; update the scene model; update positions of moving objects; finding a set of descriptors that encapsulate the information necessary to reconstruct a frame; replace parts of the frame occupied by the objects marked for remove with use of a 3D patches. In this paper, we extend an image inpainting algorithm based texture and structure reconstruction by incorporating an improved strategy for video. Our algorithm is able to deal with a variety of challenging situations which naturally arise in video inpainting, such as the correct reconstruction of dynamic textures, multiple moving objects and moving background. Experimental comparisons to state-of-the-art video completion methods demonstrate the effectiveness of the proposed approach.