선택한 단어 수는 입니다.
최소 단어 이상 선택하여야 합니다.
최소 단어 이상 선택하여야 합니다.
선택한 단어 수는 30입니다.
최대 10 단어까지만 선택 가능합니다.
최대 10 단어까지만 선택 가능합니다.
제안 방법
- It shows that sumRDOQ tends to become zero when sumSQ ≤ 2. In order to better imitate the output optimized level after RDOQ process, the observations from the above analysis motivates us to setup a larger quantization dead-zone when the CG is likely to become all zero after RDOQ process.
- Five sequences are tested under RA-Main coding configuration, and four QP values (22, 27, 32, and 37) are used following the common test condition [5]. The Bjontegaard Delta Bit-Rate (BDBR) is utilized to evaluate the coding performance of the proposed method against original HEVC scalar quantization with RDOQ-Off as the anchor. The BDBR performance is illustrated in Table 1, from where we can see the proposed adaptive scalar quantization scheme achieves 4.
- An adaptive scalar quantization scheme for video coding is addressed in this paper. The proposed method adaptively adjust the dead-zone interval based on the sum of the transform coefficients of each CG inside each transform block. The experiments are carried out using HEVC reference software, and the test results show that the proposed quantization scheme provides 4.
- This paper proposes an adaptive quantization scheme which adjusts the dead-zone of SQ based on the slice type and the property of coefficient groups (CG) inside each transform block (TB).
- To better approximate the output of RDOQ process behavior, we carried out a statistical analysis on the properties of each CG inside TB to understand more about the RDOQ behavior. The data used for analysis are generated from 32 frames of sequence BasketballDrive encoded under random access (RA-Main) coding configuration with quantization parameter (QP) of 22 and 32.
- To evaluate the performance of our proposed quantization scheme, the experiments are done using HEVC reference software HM 16.15 [4] with RDOQ tool disabled (RDOQ-Off). Five sequences are tested under RA-Main coding configuration, and four QP values (22, 27, 32, and 37) are used following the common test condition [5].
대상 데이터
- To better approximate the output of RDOQ process behavior, we carried out a statistical analysis on the properties of each CG inside TB to understand more about the RDOQ behavior. The data used for analysis are generated from 32 frames of sequence BasketballDrive encoded under random access (RA-Main) coding configuration with quantization parameter (QP) of 22 and 32. We calculate the sum of nonzero quantized levels inside each CG after SQ and RDOQ (denote as sumSQ and sumRDOQ, respectively).
※ AI-Helper는 부적절한 답변을 할 수 있습니다.