Home

  TD/458

UNCTAD XIII ROUND TABLE 4, ADDRESSING PERSISTENT AND EMERGING DEVELOPMENT CHALLENGES : AN INTEGRATED APPROACH TO PROMOTING STRUCTURAL TRANSFORMATION : ISSUES NOTE / PREPARED BY THE UNCTAD SECRETARIAT

Score: 439699.95 - daccess-ods.un.org/acce....nsf/get?open&DS=TD/458&Lang=E

  Virtual forum on "The Role of Standards in Accelerating Digital Transformation for Cities and Commun

Virtual forum on "The Role of Standards in Accelerating Digital Transformation for Cities and Communities" Committed to connecting the world Search for: ITU General Secretariat Radiocommunication Standardization About ITU-T Events All Groups Standards Resources BSG Study Groups Regional Presence Join ITU-T Development ITU Telecom Members' Zone Join ITU Virtual forum on "The Role of Standards in Accelerating Digital Transformation for Cities and Communities" Rollup Image You are here ITU > Home > ITU-T > ITU-T and Climate Change > Virtual forum on "The Role of Standards in Accelerating Digital Transformation for Cities and Communities" Share Page Content 10 ​​​​​​​​​​​​​ The International Telecommunication Union (ITU) organized the Virtual Forum on “The Role of Standards in Accelerating Digital Transformation for Cities and Communities" . This forum took place virtually from 13h30-16h00 CEST on 23 April 2021 .  ​Recognizing the transformative potential of digital technologies and the urgency to develop a shared vision on digital transformation, the objective of this virtual forum was to explore the role of international standards and good practices for developing, driving and accelerating digital transformation in cities and communities. (...) Enabling a city-wide digital transformation requires understanding the needs of all stakeholders, spanning consumers, corporates, SMEs, and policymakers.

Score: 439677.54 - https://www.itu.int/en/ITU-T/c...matechange/Pages/20210422.aspx

  Joint Video Team (MPEG+ITU) Document

Otherwise ( transform_size_flag equals zero), the existing 4x4 transform is used. Otherwise (the macroblock has at least one sub-partition smaller than 8x8), the existing 4x4 transform is used. 3.3 8x8 Transform The 2-D forward 8x8 transform is computed in a separable way as a 1-D horizontal (row) transform followed by a 1-D vertical (column) transform, where the corresponding 1-D transforms are given by the matrix 8 1 3 6 10 12 12 10 6 3 4 8 8 4 4 8 8 4 6 12 3 10 10 3 12 6 8 8 8 8 8 8 8 8 10 3 12 6 6 12 3 10 8 4 4 8 8 4 4 8 12 10 6 3 3 6 10 12 8 8 8 8 8 8 8 8 × ú ú ú ú ú ú ú ú ú ú ú û ù ê ê ê ê ê ê ê ê ê ê ê ë é - - - - - - - - - - - - - - - - - - - - - - - - - - - - as proposed in [2]. (...) The number of operations required for the 8x8 2-D inverse transform of one 8x8 block is comparable to the number of operations required for performing the 2-D inverse transform of the corresponding four 4x4 blocks using a fast butterfly implementation of the existing 4x4 inverse transform [5]. (...) Gordon, “Simplified Use of 8x8 Transform”, Doc. JVT-I022, San Diego, Sept. 2003. [4] S.

Score: 439538.13 - https://www.itu.int/wftp3/av-a...te/2004_03_Munich/JVT-K028.doc

  ITU-T Video Coding Experts Group Document

Another solution could be to use a similar integer transform as for the 4x4 transform. The integer transform simplifies the exact definition of inverse transform. The present document gives information of a couple of possible integer 8 point transforms. Another issue in connection with the use of 8x8 transform, is the use of the ULVC of TML-1 for coefficient coding. (...) Since the same kinds of numbers as for the 4x4 transform are used, quantization and dequantization may be done very similarly to the 4x4 transform.

Score: 439499.07 - https://www.itu.int/wftp3/av-a...video-site/9910_Red/q15i39.doc

  Webinar: What works for gender equality and women's empowerment – a conceptual framework for evaluat

In its preamble, the "Outcome Document" calls for bold and transformative steps which are urgently needed to shift the world onto a sustainable and resilient path, and it includes a number of goals that are of transformative nature. (...) But what does "transformative change" mean in the context of the Sustainable Development Goals (SDGs)? (...) A common characteristic of transformative approaches is that they challenge existing social norms and the distribution of power and resources.

Score: 439497.75 - https://www.ifad.org/en/web/io...uating-transformational-change

  E/ECA/CM/47/2

INDUSTRIALIZATION FOR INCLUSIVE AND TRANSFORMATIVE DEVELOPMENT IN AFRICA

Score: 439358.15 - https://daccess-ods.un.org/acc...t?open&DS=E/ECA/CM/47/2&Lang=E

  ST/SG/AC.10/C.3/2019/38

TRANSPORT OF TRANSFORMERS WITH GAS CYLINDERS

Score: 439355.6 - https://daccess-ods.un.org/acc...ST/SG/AC.10/C.3/2019/38&Lang=E

  ITU-T Video Coding Experts Group Document

However, it is better to do inverse transform first for lowest possible dynamic precision during inverse transform computations. 6 2x2 chroma DC transform and quantization Forward transform Chroma DC transform is added on top of chroma transform. (...) Table 2 – Precision Luma and chroma transform and quantization Input bits Output bits Forward transform 9, 11, 13 15, 16, 16 Quantization 15, 16, 16 10, 12, 14 Dequantization 10, 12, 14 15, 17, 19 Inverse transform 15, 17, 19 9, 11, 13 Luma DC transform and quantization Input bits Output bits Transform 13, 15, 16 16, 16, 16 Quantization 16, 16, 16 12, 14, 16 Inverse transform 12, 14, 16 12, 14, 16 Dequantization 12, 14, 16 15, 17, 19 Chroma DC transform and quantization Input bits Output bits Transform 13, 15, 16 15, 16, 16 Quantization 15, 16, 16 11, 13, 15 Inverse transform 11, 13, 15 11, 13, 15 Dequantization 11, 13, 15 15, 17, 19 Since the propose transform can be implemented using only additions and shifts the multiplications are only used during quantization and de-quantization. (...) Table 3 – Encoder operations 4x4 transform for 24 4x4 luminace and chrominance blocks Process Operation Number per block 4x4 forward transform + 64 << 16 // 0,4 or 16 4x4 transform for one 4x4 luminace DC block Process Operation Number per block 4x4 forward transform + 64 // 16 2x2 transform for two 2x2 Chrominace DC blocks Process Operation Number per block 4x4 forward transform + 8 // 0 or 4 quantization for coefficient Process Operation Number per coefficient Quantization * 1 // 1 Table 4 – Decoder operations 4x4 inverse transform and dequantization for 24 4x4 luminace and chrominance blocks Process Operation Number per block Dequantization * 16 4x4 inverse transform + 64 >> 16 Normalization + 16 >> 4x4 inverse transform and dequantization for one 4x4 luminace DC block Process Operation Number per block 4x4 inverse transform + 64 Dequantization * 16 // 16 2x2 inverse transform and dequantization for two 2x2 Chrominace DC blocks Process Operation Number per block 2x2 inverse transform + 8 Dequantization * 4 // 4 8 Results The attached document contains results when the encoder is implemented to ensure only 16 bit multiplications for 9, 11 and 13 bit residual data.

Score: 439298.2 - https://www.itu.int/wftp3/av-a...e/2001_12_Pattaya/VCEG-O25.doc

  ITU-T Video Coding Experts Group Document

However, it is better to do inverse transform first for lowest possible dynamic precision during inverse transform computations. 6 2x2 chroma DC transform and quantization Forward transform Chroma DC transform is added on top of chroma transform. (...) Table 2 – Precision Luma and chroma transform and quantization Input bits Output bits Forward transform 9, 11, 13 15, 16, 16 Quantization 15, 16, 16 10, 12, 14 Dequantization 10, 12, 14 15, 17, 19 Inverse transform 15, 17, 19 9, 11, 13 Luma DC transform and quantization Input bits Output bits Transform 13, 15, 16 16, 16, 16 Quantization 16, 16, 16 12, 14, 16 Inverse transform 12, 14, 16 12, 14, 16 Dequantization 12, 14, 16 15, 17, 19 Chroma DC transform and quantization Input bits Output bits Transform 13, 15, 16 15, 16, 16 Quantization 15, 16, 16 11, 13, 15 Inverse transform 11, 13, 15 11, 13, 15 Dequantization 11, 13, 15 15, 17, 19 Since the propose transform can be implemented using only additions and shifts the multiplications are only used during quantization and de-quantization. (...) Table 3 – Encoder operations 4x4 transform for 24 4x4 luminace and chrominance blocks Process Operation Number per block 4x4 forward transform + 64 << 16 // 0,4 or 16 4x4 transform for one 4x4 luminace DC block Process Operation Number per block 4x4 forward transform + 64 // 16 2x2 transform for two 2x2 Chrominace DC blocks Process Operation Number per block 4x4 forward transform + 8 // 0 or 4 quantization for coefficient Process Operation Number per coefficient Quantization * 1 // 1 Table 4 – Decoder operations 4x4 inverse transform and dequantization for 24 4x4 luminace and chrominance blocks Process Operation Number per block Dequantization * 16 4x4 inverse transform + 64 >> 16 Normalization + 16 >> 4x4 inverse transform and dequantization for one 4x4 luminace DC block Process Operation Number per block 4x4 inverse transform + 64 Dequantization * 16 // 16 2x2 inverse transform and dequantization for two 2x2 Chrominace DC blocks Process Operation Number per block 2x2 inverse transform + 8 Dequantization * 4 // 4 8 Results The attached document contains results when the encoder is implemented to ensure only 16 bit multiplications for 9, 11 and 13 bit residual data.

Score: 439298.2 - https://www.itu.int/wftp3/av-a...2001_12_Pattaya/VCEG-O25r1.doc

  ITU-T Video Coding Experts Group Document

However, it is better to do inverse transform first for lowest possible dynamic precision during inverse transform computations. 6 2x2 chroma DC transform and quantization Forward transform Chroma DC transform is added on top of chroma transform. (...) Table 2 – Precision Luma and chroma transform and quantization Input bits Output bits Forward transform 9, 11, 13 15, 16, 16 Quantization 15, 16, 16 10, 12, 14 Dequantization 10, 12, 14 15, 17, 19 Inverse transform 15, 17, 19 9, 11, 13 Luma DC transform and quantization Input bits Output bits Transform 13, 15, 16 16, 16, 16 Quantization 16, 16, 16 12, 14, 16 Inverse transform 12, 14, 16 12, 14, 16 Dequantization 12, 14, 16 15, 17, 19 Chroma DC transform and quantization Input bits Output bits Transform 13, 15, 16 15, 16, 16 Quantization 15, 16, 16 11, 13, 15 Inverse transform 11, 13, 15 11, 13, 15 Dequantization 11, 13, 15 15, 17, 19 Since the propose transform can be implemented using only additions and shifts the multiplications are only used during quantization and de-quantization. (...) Table 3 – Encoder operations 4x4 transform for 24 4x4 luminace and chrominance blocks Process Operation Number per block 4x4 forward transform + 64 << 16 // 0,4 or 16 4x4 transform for one 4x4 luminace DC block Process Operation Number per block 4x4 forward transform + 64 // 16 2x2 transform for two 2x2 Chrominace DC blocks Process Operation Number per block 4x4 forward transform + 8 // 0 or 4 quantization for coefficient Process Operation Number per coefficient Quantization * 1 // 1 Table 4 – Decoder operations 4x4 inverse transform and dequantization for 24 4x4 luminace and chrominance blocks Process Operation Number per block Dequantization * 16 4x4 inverse transform + 64 >> 16 Normalization + 16 >> 4x4 inverse transform and dequantization for one 4x4 luminace DC block Process Operation Number per block 4x4 inverse transform + 64 Dequantization * 16 // 16 2x2 inverse transform and dequantization for two 2x2 Chrominace DC blocks Process Operation Number per block 2x2 inverse transform + 8 Dequantization * 4 // 4 8 Results The attached document contains results when the encoder is implemented to ensure only 16 bit multiplications for 9, 11 and 13 bit residual data.

Score: 439298.2 - https://www.itu.int/wftp3/av-a...deo-site/0112_Pat/VCEG-O25.doc