KEYNOTE SPEAKER

Timothy K. Shih
Multimedia Information Network Lab
National Central University, Taiwan
E-mail: TimothyKShih@gmail.com

Video Forgery and Motion Editing

Abstract

Video Forgery is a technique for generating fake video by altering, combining, or creating new video contents. We change the behavior of actors in a video. For instance, the outcome of a 100-meter race in the Olympic Game can be falsified. We track objects and segment motions using a modified mean shift mechanism. The resulting video layers can be played in different speeds and at different reference points with respect to the original video. In order to obtain a smooth movement of target objects, a motion interpolation mechanism is proposed based on reference stick figures (i.e., a structure of human skeleton) and video inpainting mechanism. The video inpainting mechanism is performed in a quasi-3D space via guided 3D patch matching. Interpolated target objects and background layers are fused. It is hard to tell whether a falsified video is the original.

In addition, in this talk, we demonstrate a new technique to allow users to change the dynamic texture used in a video background for special effect production. For instance, the dynamic texture of fire, smoke, water, cloud, and others can be edited through a series of automatic algorithms. Motion estimations of global and local textures are used. Video blending techniques are used in conjunction with a color balancing technique. The editing procedure will search for suitable patches in irregular shape blocks, to reproduce a realistic dynamic background, such as large waterfall, fire scene, or smoky background. The technique is suitable for making science fiction movies.

We demonstrate the original and the falsified videos in our website at http://www.csie.ncu.edu.tw/~tshih. Although video falsifying may create a moral problem, our intension is to create special effects in movie industry.

Keywords: Video Forgery, Object Tracking, Motion Estimation, Special Effect Production

KEYNOTE SPEAKER

Prof. Yuji Oie
Vice President
Kyushu Institute of Technology Japan
E-mail: oie@cse.kyutech.ac.jp

Networks with Diversity and Their Technologies

Abstract

The network plays a vital role in our society and is expected to evolve to meet the changing demands and constraints. The demands for the use of networks are diverse so that they are used by diverse applications and in diverse environments. In particular, the number of mobile users of the networks significantly increases, so that much more wireless frequency will be needed in near future. The wireless frequencies are diverse, in nature, in terms of their transmission capacity and coverage.

Almost all of them are already assigned to specific use and are not available except for predetermined use, while their utilization varies in time and can be sometime quite limited. In addition, mobile users can be in diverse environment: sometimes in challenged environment which can be featured by huge communication error or/and intermittent links. In this talk, I will focus on two networks: Dynamic Spectrum Access Network (DSAN) and Delay, Disruption, Disconnection Tolerant Network (DTN). The DSAN enables effective use of diverse frequency, while the assigned frequency is not always used efficiently yet.

We have conducted vehicle communications field tests in Japan using TV white space and I will show a video about the field tests. The DTN enables communication in challenged environments. Furthermore, performance metric used for network design becomes diverse; e.g., energy saving is now regarded as important one, which leads to green networking. Network topologies are also diverse; e.g., nodes are not always uniform, but some of them are linked with a great number of nodes, while others are not. These features and networking technologies will be addressed as well.