Grant: Strategic Projects - Group

The recent advances in high dynamic range (HDR) display devices promise to provide life-like picture quality, thanks to an improved luminance/color dynamic range over the existing conventional display technology. However, for these displays to widely penetrate the consumer market, it is necessary that other supporting technologies are also developed. The goal of this project is to develop a complete HDR video capturing, compression and transmission solution, which will support and complement the emerging HDR display technology and thus enable the HDR market. The novel aspects of our research are explained below: 1.Presently the market is dominated by Low Dynamic Range (LDR) display devices. Our proposed research will enable the generation of LDR backward compatible video from HDR content. Our generated LDR content should offer picture quality that is superior to that captured by existing LDR devices and traditional HDR to LDR conversion methods. Unlike existing methods, we will develop a compression-aware HDR-to-LDR tone-mapping algorithm that will jointly minimize the overall quantization losses, resulting from both tone-mapping and compression. 2.The HDR video content will be coded in a scalable video format that maximizes the picture quality and efficiently encodes each layer so that the overall bandwidth requirements are kept within industry-acceptable limits. This method will produce backward compatible LDR component and allow perfect reproduction of the HDR video. 3.At present, HDR video cameras are extremely expensive. To facilitate early market adoption of HDR technology, our research will provide highly cost-effective HDR video capturing solutions using regular LDR cameras. Unlike existing complicated approaches that use one LDR camera and suffer from temporal artifacts, we propose to use two synchronized LDR cameras to eliminate these artifacts and produce better HDR picture quality.