报告题目：Digital Image and Video Processing Using Subpixel Rendering
Often a high-resolution RGB color image/video needs to be displayed on a low-resolution LCD display. For example, a 10 mega-pixel image is to be displayed on an SVGA monitor, or a 4CIF image is to be displayed on a QCIF mobile phone display. Limited by the low-resolution display, we have to shrink the image/video from high-resolution to low-resolution. Signal processing theory tells us that optimal decimation requires low-pass filtering with a suitable cut-off frequency followed by down-sampling. In doing so, we need to remove lots of details in the original high-resolution image.
An interesting question is – Can we display the image with higher “resolution” than what the low-resolution display has? In other words, can we get more than one pixel “resolution” out of one pixel? The answer is a surprising “yes, but …”. In this talk, we review some little known results on this interesting topic called subpixel rendering and attempt to explain what happens and why this is even possible. It can provide apparent higher resolution at the price of some, often annoying, color artifacts.
Lu Fang received the Bachelor’s degree from University of Science and Technology of China (USTC) in 2007 and the Ph.D degree from Hong Kong University of Science and Technology (HKUST) in 2011. She used to visit Northwestern University (USA) with the host of Prof. Aggelos K. Katsaggelos in 2010. Now she is a research associate in the Department of Electrical and Computer Engineering at Hong Kong University of Science and Technology. Her current research interests are coding and processing of both image and video, in particular, subpixel rendering, demosaicing, HDR imaging, compressive sensing.