學術論文

    Eyes Wide Open

    Friday, 22 November

    16:15 - 18:00

    Convention Hall B

    Near-Eye Light Field Displays

    Near-eye light field displays depict sharp images by synthesizing light fields corresponding to virtual scenes located within a viewer's natural accommodation range. We establish fundamental trade-offs between resolution, field of view, and form factor and demonstrate a thin, lightweight HMD prototype, containing a pair of microlens-covered OLEDs.


    Douglas Lanman, NVIDIA Research
    David Luebke, NVIDIA Research

    Joint View Expansion and Filtering for Automultiscopic 3D Displays

    We propose the first technique that performs both view expansion from 3D stereo content to multiple views and antialiasing for automultiscopic 3D displays. Our method can be efficiently implemented on GPU and provide results that are favorable when compare to state-off-the-art techniques,


    Piotr Didyk, Massachusetts Institute of Technology
    Pitchaya Sitthi-Amorn, Massachusetts Institute of Technology
    William Freeman, Massachusetts Institute of Technology
    Fredo Durand, Massachusetts Institute of Technology
    Wojciech Matusik, Massachusetts Institute of Technology

    A Metric of Visual Comfort for Stereoscopic Motion

    We propose a novel metric of visual comfort for stereoscopic motion, based on perceptual experiments. We take into account disparity, motion in depth, motion on the screen plane, and the spatial frequency of luminance contrast. We further derive a comfort metric to predict the comfort of short stereoscopic videos.


    Song-Pei Du, Tsinghua University
    Belen Masia, University of Zaragoza
    Shi-Min Hu, Tsinghua University
    Diego Gutierrez, University of Zaragoza

    Stereoscopizing Cel Animations

    We propose a method to convert 2D cel animations into stereoscopic cel animations. It is mainly automatic with an optional user control. It fits nicely into the production flow of traditional cel animation. No 3D modeling nor manual depth assignment is required.


    Xueting Liu, The Chinese University of Hong Kong
    Xiangyu Mao, The Chinese University of Hong Kong
    Xuan Yang, The Chinese University of Hong Kong
    Linling Zhang, The Chinese University of Hong Kong
    Tien-Tsin Wong, The Chinese University of Hong Kong