Optimized Lossless Embedded Compression for Mobile Multimedia Applications

Abstract

Power consumption is a critical design factor in modern mobile chip design, in which the memory system with dynamic random-access memory (DRAM) consumes more than half of the entire system’s power. Without DRAM bandwidth compression, extreme multimedia operations such as 8K high dynamic range (HDR) recording and 8K video conference calling are not possible without sacrificing image quality or trimming because of thermal limitations or battery time sustainability constraints. Since heterogeneous processors are substantially involved in managing various types of fallbacks or software solutions, complicated compression algorithms for high-compression ratios are not actually adaptable owing to timing closure problems or high throughput requirements. In this paper, we propose evaluation metrics to assess lossless embedded compression (LEC) algorithms to reflect realistic design considerations for mobile multimedia scenarios. Furthermore, we introduce an optimized LEC implementation for contemporary multimedia applications in mobile devices based on the proposed metrics. The proposed LEC implementation enhances the compression ratio of LEC algorithms in other commercial application processors for contemporary premium smartphones by up to 9.2% on average, while maintaining the same timing closure condition.

A Lossless Intra Reference Block Recompression Scheme for Bandwidth Reduction in HEVC-IBC

“…Table Ⅲ reveals that both the proposed scheme and MED prediction [11,12] can reach the highest DRR of 72%. Compared with the DIP [9] and GAP [10,12], the proposed scheme achieves a better DRR with lower complexity since seven neighboring samples are utilized to model the content in DIP and GAP, while our proposal takes only three samples.…”

“…To evaluate the efficiency of the proposed Intra-RBRC scheme, especially the prediction part, six previous prediction works are also simulated on SVO-VLC, including horizontal DPCM prediction (HD) in [7], both horizontal and vertical DPCM prediction (HVD) with one extra signaling bit encoded in [8], a modified version of In-Block Prediction (4IBP) [4] that use the frequently used mode 0, 4, 5, and 6 instead of the intra mode mapping, Self-Contained Directional Intra Prediction (DIP) in [9], gradient-adjusted prediction (GAP) in [10,12], and median edge detector (MED) in [11,12]. The simulation results on the unfiltered reconstruction frame used by IBC are shown in Table Ⅲ.…”

“…GAP uses 7 samples to model the texture information and MED uses 3 samples. [12] compares the performance of GAP and MED in embedded compression.…”

Lossless Decompression Accelerator for Embedded Processor with GUI

“…With the development of the mobile industry, the user-centered application market has increased significantly and, as a result, the embedded system has become high-performance to meet advanced users’ requirements [ 1 ]. In particular, the embedded system is equipped with a display in order to provide a visible environment to a user interface and attempts to achieve a high-performance graphical user interface (GUI) in the embedded system have appeared [ 2 ].…”

“…In addition, an optimized system design is possible because of the area benefit generated by utilizing compressed data. Several compression algorithms have been researched to apply a compression accelerator on the embedded system [ 1 , 11 ]. The lossy compression algorithm has been dominant owing to the relatively high compression ratio [ 8 ].…”