A Highly Parallel Architecture for Deblocking Filter in H.264/AVC

Li, Lingfeng; Goto, Satoshi; Ikenaga, Takeshi

doi:10.1093/ietisy/e88-d.7.1623

Cited by 27 publications

(38 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposal of Li et al [6] is based on both data reuse and concurrence principles. The processing order, shown in Figure 6, reduces significantly the number of cycles needed to filter a macroblock through the execution of horizontal and vertical filterings at the same time.…”

Section: Previously Proposed Filtering Ordersmentioning

confidence: 99%

“…Similarly to Li et al [6], Jing et al [7] developed a design with two edge-filters that process the vertical and horizontal edges in parallel. The difference between them is that Jing's order [7], presented in Figure 7, starts the parallel processing in the fourth state instead of the third.…”

Section: Previously Proposed Filtering Ordersmentioning

confidence: 99%

“…The difference between them is that Jing's order [7], presented in Figure 7, starts the parallel processing in the fourth state instead of the third. This difference allows the use of both filters continuously, differently from Li's order [6], which performs only vertical filterings in stages 5, 10, 15, 20, and 21.…”

Section: Previously Proposed Filtering Ordersmentioning

confidence: 99%

“…Also based on Li's proposal [6], a concurrent architecture that uses four filtering cores is presented by Ernst [10]. However, as in all the previous block-based orders presented, the number of concurrent filterings is still very limited due to the data dependencies between the blocks.…”

Section: Previously Proposed Filtering Ordersmentioning

confidence: 99%

“…48 * 4T = 192T 1 256 bytes Li et al [6] 3 5 * 4T = 140T 2 112 bytes Jing et al [7] 2 7 * 4T = 108T 2 640 bytes Zhou et al [8] 32 * 4T = 128T 2 579 bytes Lin et al [9] 2 4 * 4T = 96T 2 416 bytes Ernst et al [10] 7 5 T 4 224 bytes This work 53T 4 128 bytes this work in terms of memory use is the one proposed in [6]. However, the number of cycles to filter a complete macroblock is almost three times higher in this case.…”

mentioning

confidence: 99%

See 4 more Smart Citations

Sample-Level Filtering Order for High-Throughput and Memory-Aware H.264 Deblocking Filter

Corrêa

Agostini

Cruz

2012

ISRN Signal Processing

View full text Add to dashboard Cite

This paper presents a new sample-level filtering order for the Deblocking Filter process of the H.264/AVC video coding standard to be used instead of the traditional block-level order presented in previous works. This processing order allows a better exploration of the parallelism in the filtering process by reducing data dependencies in comparison to other works. The proposed samplelevel order allows four parallel and independent samples filtering simultaneously, completing one complete macroblock filtering in fewer cycles and requiring less memory space than the related works. The proposed filtering order can be applied to the Deblocking Filter presented in a conventional H.264/AVC encoder or decoder and to the H.264/SVC interlayer Deblocking Filter. When compared to the original H.264/AVC filter and to the best related work found in the literature, the proposed scheme achieves a reduction of 72% and 25% in the number of clock cycles and a memory usage decrease of 75% and 43%, respectively.

show abstract

Section: Previously Proposed Filtering Ordersmentioning

confidence: 99%