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

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

Primary tabs

Citation Author(s):
Sergey Zvezdakov, Dmitriy Vatolin
Submitted by:
Roman Kazantsev
Last updated:
24 April 2020 - 2:47pm
Document Type:
Poster
Document Year:
2020
Event:
Presenters Name:
Roman Kazantsev
Paper Code:
171

Abstract 

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:

Dataset Files

Kazantsev_Zvezdakov_Vatolin_poster.pdf

(217)