Image matting methods based on deep learning have made tremendous success. However, the success of previous image typically relies a massive amount pixel-level labeled data, which are time-consuming and costly to obtain. This paper first proposes semi-supervised algorithm semantic consistency trimaps (Tri-SSL), uses provide weakly supervised signals for unlabeled reduce labeling cost. Tri-SSL i...

We shall demonstrate a system that performs live-action compositing of physical and virtual objects against a panoramic background image in real-time at interactive rates. A static camera is directed towards a 40 cm3 miniature stage, whose backdrop has been colored in chromatic green. Users can add virtual objects and manipulate their parameters within the scene by using a proxy device that con...

In this paper, we propose to rely on a recent image representation model, namely the α-tree, to achieve efficient segmentation of images and videos. The α-tree is a multiscale representation of an image, based on its quasi-flat zones. An in-depth study of this tree reveals some interesting features of image pixels and regions. These features are then used in the design of both automatic and int...

Digital compositing is an important topic in the field of computer graphics and image processing. Many works have sought a good method to prevent composited images from being distinguishable from real images. The paper proposes a new environment matting algorithm to model the appearance of transparent object under different backgrounds. In the proposed approach, the frequency-domain analysis is...

We present a novel approach to real-time spatial rendering of realistic auditory environments and sound sources recorded live, in the field. Using a set of standard microphones distributed throughout a real-world environment, we record the sound field simultaneously from several locations. After spatial calibration, we segment from this set of recordings a number of auditory components, togethe...