Rate-Distortion-Complexity Optimization of an H.264/AVC Encoder for Real-Time Videoconferencing on a Mobile Device

This paper presents an H.264 standard-compliant video encoder optimized for region-of-interest (ROI) based coding and applied to American Sign Language (ASL) videos. Three encoding parameters are developed which allow the encoder to allocate both rate and computational resources differently between the ROI and non-ROI. An objective measure of intelligibility is included in an encoder parameter optimization by modifying a fast offline distortion-complexity optimization algorithm, resulting in parameter selections that demonstrate excellent rate-distortion-complexity performance. These parameters can be stored in a look-up table for use by an online algorithm that selects parameters based on available computational resources. The offline training is performed both on a PC and a cell phone. The resulting parameter selections improve the encoder speed by up to 54.4% on the PC and 62.1% on the cell phone with a small decrease in intelligibility over the x264 default parameter setting.