Tech-invite 3GPPspace IETFspace

21 22 23 24 25 26 27 28 29 31 32 33 34 35 36 37 38 4‑5x

Content for TR 26.855 Word version: 19.0.0

0… 4…

4 Compressibility analysis of film grain test sequences p. 7

4.1 Overview p. 7

4.2 Encoding configurations p. 8

4.3 Evaluation results and conclusion p. 9

5 Film grain characteristics SEI message capabilities and tuning tool p. 9

5.1 Introduction p. 9

5.2 FGC SEI message designer p. 9

5.3 Description of the tool p. 9

5.3.1 Graphical display and edit p. 9

5.3.2 Example configuration import / export p. 10

5.3.3 Tool chain for experimentation p. 11

6 Interoperability of the FGC SEI message p. 12

6.1 Film Grain Characteristics SEI message p. 12

6.2 Use of the FGC SEI message with a fixed grain pattern implementation p. 15

7 Potential additions to TR 26.955 for FGS scenario p. 15

8 Considerations on UE requirements p. 20

9 Communication with JVET p. 20

10 Conclusion p. 20

A Compressibility analysis of Film Grain - test results p. 21

A.1 Simulation Results p. 21

B Examples of use of the FG SEI cap and tuning tool p. 51

B.1 Example of use p. 51

B.1.1 Prepare original and degrained/decoded videos p. 51

B.1.2 Select a decoded picture for study p. 51

B.1.3 Focus on a light area, tune amplitude and size p. 53

B.1.4 Split interval and focus on a darker area p. 54

B.1.5 Create more intervals, and repeat on chroma p. 56

B.2 Concluding remarks p. 56

C Preliminary test results p. 58

C.1 FGS for preserving artistic intent p. 58

C.1.1 Workflow p. 58

C.1.2 Quality Evaluation Methods for FGS p. 59

C.1.3 Subjective Performance Evaluation Methods p. 59

C.1.4 Scenario FHD Test Results (Subjective) p. 60

C.1.4.1 Disclaimer on the results documented p. 60

C.1.4.2 BQTerrace Sequence p. 60

C.1.4.3 OldTownCross Sequence p. 61

C.1.4.4 InToTree Sequence p. 61

C.1.5 Scenario 4K-TV Test Results (Subjective) p. 62

C.1.5.1 CrowdRun Sequence p. 62

C.1.5.2 Tears of Steel Sequence p. 62

C.1.6 Summary remarks from proponent for FGS subjective test results for preserving artistic intent p. 66

C.2 FGS for masking coding artifacts p. 67

C.2.1 Workflow p. 67

C.2.2 Subjective Test Results A|B Method with DMOS p. 68

C.2.3 Just Noticeable Difference (JND) subjective test method p. 68

D Encoding configuration parameters p. 69

$ Change history p. 74