Use of color is a central part of video analysis, necessitating a clear understanding of the space of color representations (e.g., RGB, HSI, YCbCr), color imaging processes, and their computational requirements. This research developed a color-aware instruction set extension that exploits subword parallelism to enhance the performance and storage efficiency in the processing of color images and video. Using a luminance-chrominance color coordinate space reduces storage requirements significantly, while maintaining image quality. The color-aware extensions have been applied to both dynamically scheduled superscalar processors and to SIMD array architectures, outperforming traditional multimedia extensions in terms of speedup, energy reduction, and area efficiency.

Publications:

• J. Kim, L. M. Wills and S. Wills, Color-Aware Instructions for Embedded Superscalar Processors, to appear Journal of Signal Processing Systems, Springer, 2010.
• J. M. Kim, S. Wills and L. M. Wills, Implementing and Evaluating a Color-Aware Instruction Set for Low-Memory, Embedded Video Processing in Data Parallel Architectures, Proceedings of the IFIP International Conference on Embedded and Ubiquitous Computing (EUC 2005), Lecture Notes in Computer Science, Vol. 3824, Springer-Verlag, pp. 4 - 16, Nagasaki, Japan, Best Paper Award, December 2005.
• J. M. Kim, S. Wills and L. M. Wills, Architectural Enhancements for Color Image and Video Processing on Embedded Systems, Proceedings of the 10th Asia-Pacific Conference: Advances in Computer Systems Architecture (ACSAC 2005), Lecture Notes in Computer Science, Vol. 3740, Springer-Verlag, pp. 104 - 117, Singapore, October 2005.
• J. M. Kim, S. Wills and L. M. Wills, Determining Optimal Grain Size for Efficient Vector Processing on SIMD Image Processing Architectures, Proceedings of the 10th Asia-Pacific Conference: Advances in Computer Systems Architecture (ACSAC 2005), Lecture Notes in Computer Science, Vol. 3740, Springer-Verlag, pp. 551 - 565, Singapore, October 2005.
• J. M. Kim, L. M. Wills and D. S. Wills, Effective Detection and Elimination of Impulse Noise for Reliable 4:2:0 YCbCr Signals Prior to Compression Encoding, Proceedings of the 30^th IEEE International Conference on Acoustics, Speech, Signal Processing (ICASSP '05), Vol. 2, pp. 1005-1008, Philadelphia, Pennsylvania, March 2005.
• J. M. Kim and D. S. Wills, Evaluating a 16-bit YCbCr (6:5:5) Color Representation for Low Memory, Embedded Video Processing, Proceedings of the IEEE International Conference on Consumer Electronics (ICCE '05), pp. 181-182, Las Vegas, Nevada, January 2005.
• J. M. Kim and D. S. Wills, Efficient Processing of Color Image Sequences using a Color-Aware Instruction Set on Mobile Systems, Proceedings of the 15^th IEEE International Conference on Application-Specific Systems, Architectures and Processors (ASAP 2004), pp. 137-149, Galveston, Texas, September 2004.
• J. M. Kim and D. S. Wills, Combining the Quantized Color Instruction Set and Loop Unrolling on Portable Video Processing Systems, Proceedings of 16^th SPIE-IS&T Electronic Imaging, Embedded Processors for Multimedia and Communications, Vol. 5309, pp. 80-91, San Jose, California, January 2004.
• J. M. Kim and D. S. Wills, High-Performance and Energy-Efficient Heterogeneous Subword Parallel Instructions, Proceedings of the IEEE International Workshop on Signal Processing Systems (SiPS03), pp. 75-80, Seoul, Korea, August 2003.
• J. M. Kim and D. S. Wills, Quantized Color Instruction Set for Media-On-Demand Applications, Proceedings of the IEEE International Conference on Multimedia & Expo (ICME’03), pp. 141-144, Baltimore, Maryland, July 2003.
• J. M. Kim and D. S. Wills, Evaluating Color Instruction Set Extension for Real-Time Vector Quantization, Proceedings of the IEEE CAMP 2003 International Workshop on Computer Architectures for Machine Perception, pp. 113-120, New Orleans, Louisiana, May 2003.
• J. M. Kim and D. S. Wills, Fast Color Image Processing Using Quantized Color Instruction Set, Proceedings of the IEEE International Conference on Information Technology: Coding and Computing (ITCC 2003), pp. 529-535, Las Vegas, Nevada, April 2003.