Intra prediction with deep learning

Raz Birman, Yoram Segal, Avishay David-Malka, Ofer Hadar

פרסום מחקרי: פרק בספר / בדוח / בכנספרסום בספר כנסביקורת עמיתים

תקציר

One fundamental component of video compression standards is Intra-Prediction. Intra-Prediction takes advantage of redundancy in the information of neighboring pixel values within video frames to predict blocks of pixels from their surrounding pixels and thus allowing to transmit the prediction errors instead of the pixel values themselves. The prediction errors are of smaller values than the pixels themselves, thus allowing to accomplish compression of the video stream. Prevalent standards take advantage of intra-frame pixel value dependencies to perform prediction at the encoder end and transfer only residual errors to the decoder. The standards use multiple "Modes", which are various linear combinations of pixels for prediction of their neighbors within image Macro-Blocks (MBs). In this research, we have used Deep Neural Networks (DNN) to perform the predictions. Using twelve Fully Connected Networks, we managed to reduce Mean Square Error (MSE) of the predicted error by up to 3 times as compared to standard modes prediction results. This substantial improvement comes at the expense of more extensive computations. However, these extra computations can be significantly mitigated by the use of dedicated Graphical Processing Units (GPUs).

שפה מקוריתאנגלית אמריקאית
כותר פרסום המארחApplications of Digital Image Processing XLI
עורכיםAndrew G. Tescher
מוציא לאורSPIE
מסת"ב (מודפס)9781510620759
מזהי עצם דיגיטלי (DOIs)
סטטוס פרסוםפורסם - 1 ינו׳ 2018
אירועApplications of Digital Image Processing XLI 2018 - San Diego, ארצות הברית
משך הזמן: 20 אוג׳ 201823 אוג׳ 2018

סדרות פרסומים

שםProceedings of SPIE - The International Society for Optical Engineering
כרך10752

כנס

כנסApplications of Digital Image Processing XLI 2018
מדינה/אזורארצות הברית
עירSan Diego
תקופה20/08/1823/08/18

ASJC Scopus subject areas

  • ???subjectarea.asjc.2500.2504???
  • ???subjectarea.asjc.3100.3104???
  • ???subjectarea.asjc.1700.1706???
  • ???subjectarea.asjc.2600.2604???
  • ???subjectarea.asjc.2200.2208???

טביעת אצבע

להלן מוצגים תחומי המחקר של הפרסום 'Intra prediction with deep learning'. יחד הם יוצרים טביעת אצבע ייחודית.

פורמט ציטוט ביבליוגרפי