GAMES Webinar 2018 -77期（Siggraph Asia 2018论文报告）| 樊庆楠（山东大学），谈建超（乔治梅森大学）
【GAMES Webinar 2018-77期（Siggraph Asia 2018论文报告）】
报告题目：Image Smoothing via Unsupervised Learning
Image smoothing represents a fundamental component of many disparate computer vision and graphics applications. In this work, we develop a unified unsupervised (label-free) learning framework that facilitates generating flexible and high-quality smoothing effects by directly learning from data using deep convolutional neural networks (CNNs). The heart of the design is the training signal as a novel energy function that includes an edge-preserving regularizer which helps maintain important yet potentially vulnerable image structures, and a spatially-adaptive Lp flattening criterion which imposes different forms of regularization onto different image regions for better smoothing quality. We implement a diverse set of image smoothing solutions employing the unified framework targeting various applications such as, image abstraction, pencil sketching, detail enhancement, texture removal and content-aware image manipulation, and obtain results comparable with or better than previous methods. Our method is extremely fast with a modern GPU (e.g, 200 fps for 1280×720 images).
Qingnan Fan is a fifth-year PhD student from Shandong University, supervised by Prof. Baoquan Chen. He received his Bachelor degree in Shandong University, 2014. His research interests lie in computer vision, computer graphics and computational photography. Previously, he has broad research experience in Tel Aviv University, Hebrew University of Jerusalem, Microsoft Research and Cambridge University to work with Prof. Daniel Cohen-Or, Dani Lischinski, Xin Tong, Gang Hua and Carola-Bibiane Schönlieb. He has strong publications during the last four-year PhD career, including two TOG papers and three CVPR/ICCV/ECCV papers which are all in first-author and include one Oral paper, specifically image smoothing via unsupervised learning (SIGGRAPH Asia 2018), real-time video segmentation (SIGGRAPH Asia 2015), reflection removal (ICCV 2017), intrinsic image decomposition (CVPR 2018 Oral) and decouple learning for image processing tasks (ECCV 2018). He also received the Presidential Scholarship of Shandong University in 2015.
报告题目：Efficient Palette-based Decomposition and Recoloring of Images via RGBXY-space Geometry
We introduce an extremely scalable and efficient yet simple palette-based image decomposition algorithm. Given an RGB image and set of palette colors, our algorithm decomposes the image into a set of additive mixing layers, each of which corresponds to a palette color applied with varying weight. Our approach is based on the geometry of images in RGBXY-space. This new geometric approach is orders of magnitude more efficient than previous work and requires no numerical optimization. We provide an implementation of the algorithm in 48 lines of Python code. We demonstrate a real-time layer decomposition tool in which users can interactively edit the palette to adjust the layers. After preprocessing, our algorithm can decompose 6 MP images into layers in 20 milliseconds.
Jianchao Tan is a Ph.D. student of Computer Science Department (CS) in George Mason University (GMU), advised by Dr. Yotam Gingold at Creativity and Graphics Lab (CraGL). He obtained Bachelor degree from Electronic Engineering and Information Science Department (EEIS) in University of Science and Technology of China (USTC) at July 2013. His undergraduate research advisor is Dr. Xuejin Chen. His research interests are image & video processing, computer graphics, and computer vision. Specifically, he is interested in the application of geometry of color spaces and image edits. He is also interested in applying deep learning techniques to solve color related problems, such as recoloring, harmonization and colorization. His work was chosen to be presented on SIGGRAPH 2018 Doctoral Consortium. And his research was recognized by Adobe Research Fellowship Finalist 2018.
GAMES主页的“使用教程”中有 “如何观看GAMES Webinar直播？”及“如何加入GAMES微信群？”的信息；