Posters

Posters

[View]

Image & Video

	Live Image Composing We have developed the algorithm which enables us to compose images by just taking two phtographies, checking a final composed image in a live-view on a camera. Masayuki Tanaka, Tokyo Institute of Technology Shunsuke Nakamura, Tokyo Institute of Technology Masatoshi Okutomi, Tokyo Institute of Technology
	Aligning a 3D headless avatar with a user's 2D face images in real-time we propose a method for aligning a headless life-size 3D avatar with a user’s 2D face images in order to create a realistic user look-alike animatable virtual avatar. Miaolong Yuan, Institute for Infocomm Research
	A First Introduction to Metamerism Art Human eye often recognizes different spectral distribution as same color. Our research aims to innovate a novel trick art using the phenomenon called metamerism. Daisuke Miyazaki, Hiroshima City University Kazuki Nakamura, Hiroshima City University Masashi Baba, Hiroshima City University Ryo Furukawa, Hiroshima City University Masahito Aoyama, Hiroshima City University Shinsaku Hiura, Hiroshima City University Naoki Asada, Hiroshima City University
	Face Replacement in Video from a Single Image This work presents a novel system to replace the face in a video with another face from only a single image. Arthur Niswar, Institute for Infocomm Research Ee Ping Ong, Institute for Infocomm Research Zhiyong Huang, Institute for Infocomm Research
	Hole Filling by Relative Background and Propagation Map The hole filling of depth image based rendering is improved by relative background and propagation map for autostereoscopic displays. Te-Mei Wang, Industrial Technology Research Institute Hsiao-Wei Chen, Industrial Technology Research Institute
	Efficient Post-Processing Antialiasing on Integrated GPUs We present an improved morphological antialiasing algorithm that runs 2.5x faster on Intel HD graphics than the previous state-of-the-art GPU implementation. The source code is provided. Alexander Reshetov, Intel Alexei Soupikov, Intel Thomas Raoux, Intel
	Arbitrarily Focused Video Using High-speed Liquid Lens We propose a method to synthesize arbitrarily focused images and even videos using differently focused image sequences captured by high-speed focusing liquid lens. Hiroki Deguchi, The University Of Tokyo Hiromasa Oku, The University Of Tokyo Masatoshi Ishikawa, The University Of Tokyo
	Texturing and Deforming Models with Casual Images Novel texturing and deforming approach for mapping the pattern and shape of a casual image to a 3D model at the same time based on an alternating least-square approach. I-Chao Shen, Academia Sinica Sheng-Jie Luo, National Taiwan University Yi Hua Wang, National Taiwan University Yu-Mei Chen, National Taiwan University Bing-Yu Chen, National Taiwan University Wen-Huang Cheng, Academia Sinica
	Real-time Contrast Preserving Decolorization We propose a fast yet effective contrast preserving decolorization method. Our implementation takes only 30ms to process an one megapixel color input, comparable with the built-in matlab rgb2gray function. Cewu Lu, The Chinese University of Hong Kong Li Xu, The Chinese University of Hong Kong Jiaya Jia, The Chinese University of Hong Kong
	Seamless Image Cloning by a Closed Form Solution of a Modified Poisson Problem We propose a fast smooth image cloning algorithm based on a modified Poisson problem. The proposed algorithm can cloning the image preserving the original source color. Masayuki Tanaka, Tokyo Institute of Technology Ryo Kamio, Tokyo Institute of Technology Masatoshi Okutomi, Tokyo Institute of Technology

Posters

Image & Video

Live Image Composing

Aligning a 3D headless avatar with a user's 2D face images in real-time

A First Introduction to Metamerism Art

Face Replacement in Video from a Single Image

Hole Filling by Relative Background and Propagation Map

Efficient Post-Processing Antialiasing on Integrated GPUs

Arbitrarily Focused Video Using High-speed Liquid Lens

Texturing and Deforming Models with Casual Images

Real-time Contrast Preserving Decolorization

Seamless Image Cloning by a Closed Form Solution of a Modified Poisson Problem