Study on Digital Video Artifacts and Quality Metrics in Electrical Engineering

Slide Note
Embed
Share

Study conducted by Alexandre Fieno in the Electrical Engineering, Mathematics, and Computer Science Department focused on analyzing the characteristics of digital video artifacts like blockiness, blurriness, and packet loss. The research aimed to understand how these artifacts impact video quality and design effective quality metrics. Various artifact strengths and combinations were examined across seven original videos to evaluate annoyance, visibility, and interactions among artifacts. The study utilized a protocol with 23 participants and a single stimulus setup to assess the impact of different artifacts on video perception and quality.

onik
onik Follow

Uploaded on Oct 09, 2024 | 0 Views

Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript

  1. Experiment 3 Summary Alexandre Fieno Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  2. Experiment: motivation To study the characteristics of the most relevant artifacts present in digital videos: Annoyance, visibility, interactions among artifacts and content dependence To understand how artifacts affect quality, it is important in order to design good artifacts metrics It focus in three different artifacts: blockiness, blurriness and packet loss These are presented in several strengths, in several combinations and for seven original videos Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  3. Experiment Artefacts: packet loss Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  4. Experiment Artefacts: blockiness Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  5. Experiment Artefacts: blurriness Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  6. Experiment Setting: Stimuli Original 720p H.264 encoding PSNR>70db, 120 Mbps MPEG-2 120 Mbps MPEG-2 120 Mbps Combined Artefacts Video sequences with combined packet loss, blockiness and blurriness artefacts Loss Ratio (0.7%, 8.1%) Blockiness and Blurriness Strength (0.4%, 0.6%) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  7. Experiment Setting: Protocol 23 participants Single Stimulus setup (ITU BT.500) 1. calibration 2. Task free pristine videos 3. Training 4. Practice Trials 5. Scoring task (140 videos) Yes/no 0 100 3. How annoying were they? 2. did you see impairments? 1. video view Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  8. Experiment Analysis: blurriness (among videos) Ref. Mean Group Blur(0.6) Blur(0.6) + Pck (0.7) Blur(0.6) + Pck (8.1) Blur(0.4) + Pck (0.7) Blur(0.4) + Pck (8.1) ParkJoy 32.74 44.61 59.43** 24.91 51.52** IntoTree 39.70 35.48 49.74 14.70** 34.30 ParkRun 58.09 53.87 57.00 21.65** 33.65** Romeo 12.74 26.57** 56.57** 11.04 51.13** Cactus 26.52 35.48 55.48** 18.17 48.48** Basketball 32.13 35.57 62.70** 28.00 63.52** Barbecue 25.26 36.35 50.65** 17.87 40.57** ** p<.05 statistically significant 70 60 50 40 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Blur(0.4) + Pck (0.7) Blur(0.6) + Pck (0.7) Blur(0.4) + Pck (8.1) Blur(0.6) + Pck (8.1) Blur(0.6) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  9. Experiment Analysis: blockiness (among videos) Ref. Mean Bloc(0.6) Bloc(0.6) + Pck (0.7) Bloc(0.6) + Pck (8.1) Bloc(0.4) + Pck (0.7) Bloc(0.4) + Pck (8.1) Group 51.96 57.48 66.74** 44.61 57.26 ParkJoy 44.43 50.17 65.09** 37.70 58.96** IntoTree 53.35 54.09 63.09 29.91** 49.17 ParkRun 45.57 52.87 67.43** 44.17 62.87** Romeo 34.61 40.43 61.57** 22.87** 57.39** Cactus 49.52 57.09 72.78** 47.30 64.52** Basketball 60.61 62.87 72.17** 42.52** 64.13 Barbecue ** p<.05 statistically significant 80 70 60 50 40 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Bloc(0.4) + Pck (0.7) Bloc(0.6) + Pck (0.7) Bloc(0.4) + Pck (8.1) Bloc(0.6) + Pck (8.1) Bloc(0.6) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  10. Experiment Analysis: packet loss (among videos) Ref. Mean Pck(8.1) Bloc(0.6) Blur(0.6) Group 38.70 51.96** 32.74 ParkJoy 49.52 44.43 39.70 IntoTree 20.52 53.35** 58.09** ParkRun 43.96 45.57 12.74** Romeo 27.78 34.61 26.52 Cactus 43.57 49.52 32.13 Basketball 41.87 60.61** 25.26** Barbecue ** p<.05 statistically significant 70 60 50 40 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Bloc(0.6) Blur(0.6) Pck(8.1) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  11. Experiment 3 End Alexandre Fieno Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  12. Experiment Analysis: blurriness (among videos) Ref. Mean Group Blur(0.6) Blur(0.6) + Pck (0.7) Blur(0.6) + Pck (8.1) ParkJoy 32.74 44.61 59.43** IntoTree 39.70 35.48 49.74 ParkRun 58.09 53.87 57.00 Romeo 12.74 26.57** 56.57** Cactus 26.52 35.48 55.48** Basketball 32.13 35.57 62.70** Barbecue 25.26 36.35 50.65** ** p<.05 statistically significant 70 60 50 40 MOS 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Videos Cactus Basketball Barbecue Blur(0.6) + Pck (0.7) Blur(0.6) + Pck (8.1) Blur(0.6) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  13. Experiment Analysis: blurriness (among videos) Ref. Mean Group Blur(0.6) Blur(0.4) + Pck (0.7) Blur(0.4) + Pck (8.1) ParkJoy 32.74 24.91 51.52** IntoTree 39.70 14.70** 34.30 ParkRun 58.09 21.65** 33.65** Romeo 12.74 11.04 51.13** Cactus 26.52 18.17 48.48** Basketball 32.13 28.00 63.52** Barbecue 25.26 17.87 40.57** ** p<.05 statistically significant 70 60 50 40 MOS 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Videos Blur(0.4) + Pck (0.7) Blur(0.4) + Pck (8.1) Blur(0.6) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  14. Experiment Analysis: blockiness (among videos) Ref. Mean Bloc(0.6) Bloc(0.6) + Pck (0.7) Bloc(0.6) + Pck (8.1) Group ParkJoy 32.74 57.48 66.74** IntoTree 39.70 50.17 65.09** ParkRun 58.09 54.09 63.09 Romeo 12.74 52.87 67.43** Cactus 26.52 40.43 61.57** Basketball 32.13 57.09 72.78** Barbecue 25.26 62.87 72.17** ** p<.05 statistically significant 80 70 60 50 MOS 40 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Videos Bloc(0.6) + Pck (0.7) Bloc(0.6) + Pck (8.1) Bloc(0.6) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  15. Experiment Analysis: blockiness (among videos) Ref. Mean Bloc(0.6) Bloc(0.4) + Pck (0.7) Bloc(0.4) + Pck (8.1) Group 51.96 44.61 57.26 ParkJoy 44.43 37.70 58.96** IntoTree 53.35 29.91** 49.17 ParkRun 45.57 44.17 62.87** Romeo 34.61 22.87** 57.39** Cactus 49.52 47.30 64.52** Basketball 60.61 42.52** 64.13 Barbecue ** p<.05 statistically significant 70 60 50 40 MOS 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Videos Bloc(0.4) + Pck (0.7) Bloc(0.4) + Pck (8.1) Bloc(0.6) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  16. Experiment Analysis: packet loss (among videos) Ref. Mean Pck(8.1) Blur(0.4) + Pck (8.1) Blur(0.6) + Pck (8.1) Group 38.70 51.52 59.43** ParkJoy 49.52 34.30** 49.74 IntoTree 20.52 33.65** 57.00** ParkRun 43.96 51.13 56.57 Romeo 27.78 48.48** 55.48** Cactus 43.57 63.52** 62.70** Basketball 41.87 40.57 50.65 Barbecue ** p<.05 statistically significant 70 60 50 40 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Blur(0.4) + Pck (8.1) Blur(0.6) + Pck (8.1) Pck(8.1) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  17. Experiment Analysis: packet loss (among videos) Ref. Mean Pck(8.1) Bloc(0.4) + Pck (8.1) Bloc(0.6) + Pck (8.1) Group 38.70 57.26** 66.74** ParkJoy 49.52 58.96 65.09** IntoTree 20.52 49.17** 63.09** ParkRun 43.96 62.87** 67.43** Romeo 27.78 57.39** 61.57** Cactus 43.57 64.52** 72.78** Basketball 41.87 64.13** 72.17** Barbecue ** p<.05 statistically significant 80 70 60 50 40 30 20 10 0 ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Bloc(0.4) + Pck (8.1) Bloc(0.6) + Pck (8.1) Pck(8.1) Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  18. Experiment Analysis: original videos (among experiments) Mean Exp. 2 Group ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Exp. 1 Exp. 3 1.36 4.07 1.93 0.00 0.14 0.36 2.14 2.81 6.06 3.50 6.13 4.50 6.44 5.00 0.00 0.70 1.30 0.00 0.43 0.83 2.30 Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  19. Experiment Analysis: packet loss (between experiments) Mean Group ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Exp. 1 Exp. 3 80.00 91.29 43.36 67.79 69.79 75.36 82.36 38.70* 49.52* 20.52** 43.96** 27.78* 43.57* 41.87* * p<.01 statistically significant ** p<.05 statistically significant Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  20. Experiment Analysis: blockiness (between experiments) Mean Group ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Exp. 2 Exp. 3 55.44 50.75 53.75 68.88 51.06 70.00 69.44 38.70** 49.52 20.52* 43.96** 27.78** 43.57* 41.87** * p<.01 statistically significant ** p<.05 statistically significant Electrical Engineering, Mathematics and Computer Science Department - EEMCS

  21. Experiment Analysis: blurriness (between experiments) Mean Group ParkJoy IntoTree ParkRun Romeo Cactus Basketball Barbecue Exp. 2 Exp. 3 38.31 37.31 47.00 23.75 21.94 28.31 21.44 38.70 49.52 20.52** 43.96** 27.78 43.57** 41.87** * p<.01 statistically significant ** p<.05 statistically significant Electrical Engineering, Mathematics and Computer Science Department - EEMCS

Related


The Role of Artifacts and Evidence in Educator Evaluation and Support

The Role of Artifacts and Evidence in Educator Evaluation and Support

benson
If you are looking for Lighting in South Windsor

If you are looking for Lighting in South Windsor

Cathleen
Best Agricultural electrician in Tokanui

Best Agricultural electrician in Tokanui

Karen1
Looking for the best Industrial Electrical Installations in Chesterton

Looking for the best Industrial Electrical Installations in Chesterton

elen
Understanding MRI Artifacts: Causes and Examples

Understanding MRI Artifacts: Causes and Examples

ruqayyah
Understanding UML Deployment Diagrams

Understanding UML Deployment Diagrams

susannah
System Engineering Division Monthly Meeting - June 23, 2022

System Engineering Division Monthly Meeting - June 23, 2022

larry
Understanding Software Testing Metrics and Tools

Understanding Software Testing Metrics and Tools

scarlett
Responsible Use of Publication Metrics in Research Assessment

Responsible Use of Publication Metrics in Research Assessment

holland
Understanding Software Measurement and Metrics in Software Engineering

Understanding Software Measurement and Metrics in Software Engineering

asmodeus
5G Video Characterization: Reference Sequences and Metrics

5G Video Characterization: Reference Sequences and Metrics

guido
Understanding Software Process and Project Metrics

Understanding Software Process and Project Metrics

channing
Analysis of Onion Routing Security and Adversary-based Metrics

Analysis of Onion Routing Security and Adversary-based Metrics

rus_n
Metrics Utilization in Federal Audit Organizations

Metrics Utilization in Federal Audit Organizations

freiberg_d
Federal Audit Practices and Metrics Evaluation

Federal Audit Practices and Metrics Evaluation

sulo990
Evaluation Metrics for IEEE 802.11-14/0107 HEW Proposal

Evaluation Metrics for IEEE 802.11-14/0107 HEW Proposal

suon919
Discoveries at M-231 Site: Artifacts and Findings

Discoveries at M-231 Site: Artifacts and Findings

aye_len
Understanding Evaluation Metrics in Machine Learning

Understanding Evaluation Metrics in Machine Learning

mattywo
Enhancing Video Streaming with FESTIVE: Fairness, Efficiency, and Stability

Enhancing Video Streaming with FESTIVE: Fairness, Efficiency, and Stability

nhai
Recommended Updates for Engineering Program of Study

Recommended Updates for Engineering Program of Study

hoge_oah
MDIC Xavier University Device Product Quality Metrics Team and Goals

MDIC Xavier University Device Product Quality Metrics Team and Goals

mjoi
Metrics and Lessons Learned for Object-Oriented Projects

Metrics and Lessons Learned for Object-Oriented Projects

leda_66

More Related Content