Hsueh-Ming   Hang   received   Ph.D.   from  Rensselaer Polytechnic Institute, Troy, NY, in 1984. From 1984 to 1991，he was with AT&T Bell Laboratories, Holmdel, NJ, and then he joined  the  Electronics   Engineering   Department  of National Chiao Tung University (NCTU), Hsinchu, Taiwan, in December 1991.He is currently the Dean of the ECE College, NCTU. He has been actively involved in the international MPEG standards since 1984 and his current research interests include   multimedia   compression, multiview image/video   processing,   and deep-learning based video processing. Dr. Hang holds 13 patents (Taiwan, US and Japan) and has published over 190 technical papers related to image compression, signal processing, and video codec architecture. He was an associate editor (AE) of the IEEE Transactions on Image Processing(1992一1994, 2008-2012) and the IEEE Transactions on Circuits and Systems for Video Technology(1997一1999). He is a co-editor and contributor of the Handbook of Visual Communications published by Academic  Press  in  1995.  He  was  an  IEEE  Circuits  and  Systems  Society Distinguished Lecturer (2014-2015) and is a Board Member of the Asia-Pacific Signal and Information Processing Association (APSIPA) (2013-2018). He is a recipient of the IEEE Third Millennium Medal and is a Fellow of IEEE and IET and a member of Sigma Xi.

Abstract: RGB-D Video Challenges (Pt 1): Acquisition and Synthesis

    3D films have a history of several decades, but only recently 3D video consumer products are gradually becoming popular. The focus of next-generation 3D research is on the so-called virtual-viewpoint (or free-viewpoint) video system. It is also an on-going standardization item in the international ITU/MPEG Standards. Typically, a densely arranged camera array is used to acquire input images and a number of virtual view pictures are synthesized at the receiver using the depth-image based rendering (DIBR) technique. Three essential components are needed for building a virtual-view system: depth estimation, data compression, and view synthesis. We first look into the depth map acquisition issue. Recent devices such as Microsoft Kinect offer an inexpensive hardware solution to this problem. However, its depth map quality needs to be improved. A virtual view at an arbitrary viewpoint, in theory, can be well synthesized using the texture and depth information of the given multiple reference views. In reality, several technical challenges exist in producing a high-quality synthesized view such as warping, blending, ghost artifact reduction, hole filling and flickering noise reduction.

Abstract: RGB-D Video Challenges (Pt 2):Scene Composition

    3D films have a history of several decades, but only recently 3D video consumer products are gradually becoming popular. The focus of next一generation 3D research is on the so-called virtual-viewpoint (or free-viewpoint) video system. It is also an on-going standardization item in the international ITU/ MPEG Standards. In this talk, we present a multiple camera structure in RGB-D video acquisition. A 2x2 array system of Kinect v2 is constructed and the virtual view synthesis results are reported. An interesting application of virtual-view system is 3D scene composition. It is an extension of the traditional chroma key technique but we now try to merge two sets of 3D vide scenes into one consistent 3D scene. Another issue is the 3D subjective vision. 2D image quality assessment and attention model have been well studied but their 3D counterparts have fewer research reports so far. We explore a bit on the 3D image/video attention model.