Sorry, you need to enable JavaScript to visit this website.

facebooktwittermailshare

Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features

Abstract: 

Modern video codecs have many compression-tuning parameters from which numerous configurations (presets) can be constructed. The large number of presets complicates the search for one that delivers optimal encoding time, quality, and compressed-video size. This paper presents a machine-learning-based method that helps to solve this problem. We applied the method to the x264 video codec: it searches for optimal presets that demonstrate 9-20% bitrate savings relative to standard x264 presets with comparable compressed-video quality and encoding time. Our method is faster upto 10 times than existing solutions.

up
0 users have voted:

Paper Details

Authors:
Sergey Zvezdakov, Dmitriy Vatolin
Submitted On:
24 April 2020 - 2:47pm
Short Link:
Type:
Poster
Event:
Presenter's Name:
Roman Kazantsev
Paper Code:
171
Session:
Posters
Document Year:
2020
Cite

Document Files

Kazantsev_Zvezdakov_Vatolin_poster.pdf

(46)

Subscribe

[1] Sergey Zvezdakov, Dmitriy Vatolin, "Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5082. Accessed: Jun. 06, 2020.
@article{5082-20,
url = {http://sigport.org/5082},
author = {Sergey Zvezdakov; Dmitriy Vatolin },
publisher = {IEEE SigPort},
title = {Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features},
year = {2020} }
TY - EJOUR
T1 - Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features
AU - Sergey Zvezdakov; Dmitriy Vatolin
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5082
ER -
Sergey Zvezdakov, Dmitriy Vatolin. (2020). Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features. IEEE SigPort. http://sigport.org/5082
Sergey Zvezdakov, Dmitriy Vatolin, 2020. Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features. Available at: http://sigport.org/5082.
Sergey Zvezdakov, Dmitriy Vatolin. (2020). "Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features." Web.
1. Sergey Zvezdakov, Dmitriy Vatolin. Machine-Learning-Based Method for Finding Optimal Video-Codec Configurations Using Physical Input-Video Features [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5082