Skip to main content
SpringerLink
Log in

How to reuse existing annotated image quality datasets to enlarge available training data with new distortion types

  • Published:
Multimedia Tools and Applications Aims and scope Submit manuscript

Abstract

There is a continuing demand for objective measures that predict perceived media quality. Researchers are developing new methods for mapping technical parameters of digital media to the perceived quality. It is quite common to use machine learning algorithms for these purposes especially deep learning algorithms, which need large amounts of data for training. In this paper, we aim towards getting more training data with recent types of distortions. Instead of doing expensive subjective experiments, we evaluate the reuse of previously published, well-known image datasets with subjective annotation. In this contribution, the procedure of mapping Mean Opinion Scores (MOS) from an already published subjectively annotated dataset with older codecs to new codecs is presented. In particular, we map from Joint Photographic Experts Group (JPEG) distortions to newer High Efficiency Video Coding (HEVC) distortions. We have used values of three different objective methods as a connection between these two different distortion types. In order to investigate the significance of our approach, subjective verification tests were designed and conducted. The design goals led to two types of experiments, i.e. Pair Comparison (PC) test and Absolute Category Rating (ACR) test, in which 40 participants provided their opinion. Results of the subjective experiments indicate that it may be possible to use information gained from older datasets to describe the perceived quality of more recent compression algorithms.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8

Similar content being viewed by others

References

  1. Abdellaoui M, Douik A (2020) Human action recognition in video sequences using deep belief networks. Traitement du Signal 37(1):37–44

    Article  Google Scholar 

  2. Ahn S (2018) Lee, S. No-Reference Video Quality Assessment based on Convolutional Neural Network and Human Temporal Behavior, vol 2018. Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA ASC

    Google Scholar 

  3. Alizadeh M, Mohammadi A, Sharifkhani M (2018) No-reference deep compressed-based video quality assessment. 2018 8th International Conference on Computer and Knowledge Engineering, ICCKE 2018, Article number 8566395, pp 130–134

  4. Bagui K, Mehra KL (2017) Convergence of binomial to normal: multiple proofs. International Mathematical Forum 12(9):368–389

    Google Scholar 

  5. Berument H, Jelassi M (2002) The fisher hypothesis: a multicountry analysis. Appl Econ 34(13):1645–1655

    Article  Google Scholar 

  6. HM reference software (16.20), https://hevc.hhi.fraunhofer.de, (2015)

  7. Hou R, Zhao Y, Hu Y, Liu H (2020) No-reference video quality evaluation by a deep transfer CNN architecture; Signal Processing: Image Communication, Volume 83 Article number 115782

  8. ITU-T Rec. P.1401 (2012) Methods, metrics and procedures for statistical evaluation qualification and comparison of objective quality prediction models

  9. ITU-T Rec. P.910 (1996) Subjective video quality assessment methods for multimedia applications

  10. Katsigiannis S, Scovell J, Ramzan N, Janowski L, Corriveau P, Saad MA, Van Wallendael G (2018) Interpreting mos scores, when can users see a dif-ference? understanding user experience differences for photo quality, vol 3

  11. Mantiuk RK, Tomaszewska A, Mantiuk R (2012) Comparison of four subjective methods for image quality assessment. Comput Graph Forum 31(8):2478–2491. https://doi.org/10.1111/j.1467-8659.2012.03188

    Article  Google Scholar 

  12. Mohan J, Nair MS (2018) Dynamic summarization of videos based on descriptors in space-time video, vol 6. Article number 8476548

  13. Multimedia signal processing group, video quality measurement tool, https://mmspg.epfl.ch/downloads/vqmt, (2013)

  14. Ninassi A, Le Callet P, Autrusseau F (2006) Pseudo no reference image qualitymetric using perceptual data hiding. Human Vision and Electronic ImagingXI 6057:146–157

    Google Scholar 

  15. Otroshi Shahreza H, Amini A, Behroozi H (2019) No-reference video quality assessment using recurrent neural networks. 5th Iranian conference on signal processing and intelligent systems, ICSPIS, Article number 9066015

  16. Pinson MH (2003) An objective method for combining multiple subjective data sets. Vis Commun Image Process 5150:583–593

    Google Scholar 

  17. Pitrey Y, Engelke U, Barkowsky M, Pépion R, Le Callet P (2011) Aligningsubjective tests using a low cost common set, QoE for Multimeda Contentsharing, irccyn contribution, https://hal.archiveso-uvertes.fr/hal-00608310

  18. Ponomarenko N, Jin L, Ieremeiev O, Lukin V, Egiazarian K, As-tola J, Kuo CCJ (2015) Image database TID2013: Peculiarities, results andperspectives. Signal Process Image Commun 30:57–77

    Article  Google Scholar 

  19. Psychophysics toolbox version 3, http://psychtoolbox.org/download, (2006)

  20. Sheikh HR, Bovik AC (2006) Image information and visual quality. Trans Img Proc 15(2):430–444. https://doi.org/10.1109/TIP.2005.859378

    Article  Google Scholar 

  21. Sheikh HR, Wang Z, Cormack L, Bovik AC (2006) Live image quality assessment database, http://live.ece.utexas.edu/research/Quality/subjective.html

  22. Spiegel MR, Schiller JJ, Srinivasan RA (2000) Schaum’s outline of theory and problems of probability and statistics, 2nd edn. McGraw-Hill, New York. Schaum’s outline

    Google Scholar 

  23. TU-R Rec. BT 500-13 (2012) Methodology for the subjective assessment of thequality of television pictures

  24. Thurstone LL (1927) Psychophysical analysis. Am J Psychol 38 (3):368–389. http://www.jstor.org/stable/1415006

    Article  Google Scholar 

  25. Varga D, Szirányi T (2019) No-reference video quality assessment via pretrained CNN and LSTM networks. SIViP 3(8, 1):1569–1576

    Article  Google Scholar 

  26. Vranješ M, Rimac-Drlje S, Grgić K (2013) Review of objective video qualitymetrics and performance comparison using different databases. Signal Process Image Commun 28(1):1–19

    Article  Google Scholar 

  27. VQEG multimedia testplan phase 1, https://www.its.bldrdoc.gov/vqeg/projects/multimedia-phase-i/multimedia-phase-i.aspx (2008)

  28. Wallace GK (1992) The JPEG still picture compression standard. IEEE Transactions on Consumer Electronics 38

  29. Wang Z, Bovik AC (2002) A universal image quality index. IEEE Signal Process Lett 9:81–84. https://doi.org/10.1109/97.995823

    Article  Google Scholar 

  30. Wang Z, Bovik AC, Sheikh HR, Simoncelli EP (2004) Image quality assessment: From error measurement to structural similarity. IEEE Trans Image Processing 13:600–612

    Article  Google Scholar 

  31. Wang C, Su L, Huang Q (2017) CNN-MR for no reference video quality assessment. Proceedings - 2017 4th International Conference on Information Science and Control Engineering, ICISCE 2017, Article number 8110283, pp 224–228

  32. You J, Korhonen J (2019) Deep neural networks for no-reference video quality assessment. Signal processing: Proceedings - International Conference on Image Processing, ICIP, Volume 2019–September, September 2019, Article number 8803395, pp 2349–2353

  33. Zhang Z, Zhau J, Liu N, Gu X, Zhang Y (2017) An improved pairwise comparison scaling method for subjective image quality assessment. In: 2017 IEEE international symposium on broadband multimedia systems and broadcasting (BMSB) pp 1–6, https://doi.org/10.1109/BMSB.2017.7986235

Download references

Author information

Authors and Affiliations

  1. Faculty of Electrical Engineering and Information Technology, University of Zilina, Univerzitna 8215/1, 010 26, Zilina, Slovakia

    Tomas Mizdos, Miroslav Uhrina & Peter Pocta

  2. Faculty of Computer Science, Deggendorf Institute of Technology, Dieter-Görlitz-Platz 1, 94469, Deggendorf, Germany

    Marcus Barkowsky

Authors
  1. Tomas Mizdos
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Marcus Barkowsky
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Miroslav Uhrina
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Peter Pocta
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Tomas Mizdos.

Additional information

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Mizdos, T., Barkowsky, M., Uhrina, M. et al. How to reuse existing annotated image quality datasets to enlarge available training data with new distortion types. Multimed Tools Appl 80, 28137–28159 (2021). https://doi.org/10.1007/s11042-021-10679-5

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11042-021-10679-5

Keywords

Search

Navigation