Figure - uploaded by Pierrick Philippe
Content may be subject to copyright.
VTM Software coding configuration.
Source publication
In 2021, a new track has been initiated in the Challenge for Learned Image Compression : the video track. This category proposes to explore technologies for the compression of short video clips at 1 Mbit/s. This paper proposes to generate coded videos using the latest standardized video coders, especially Versatile Video Coding (VVC). The objective...
Context in source publication
Context 1
... Rate Distorsion optimization provides considerable performance improvement for VVC compared to a configuration where all the sequences use the same quantization parameter. Roughly 40% bit rate saving is noticed from HM to 16 17 18 19 20 21 50 60 70 80 90 100 110 120 130 140 150 -10 x log10(1-MS-SSIM) Submission size relative to challenge limit (%) VTM (all) VTM VTM no RDO HM x265 Figure 3. Performance of VTM with competition VVC and from x265 in its default configuration relative to HM. ...Citations
... For an introduction on the Versatile Video Coding standard, the reader should refer on [1] to have an overview of VVC and its development phase. Also, in 2021, VVC was also contributed as anchors for the CLIC challenge, more details can be found in [3]. ...
... Note that this optimization method was already chosen last year for the generation of VVC anchors [3] : the results of last year's challenge using this metric was shown consistent with the MS-SSIM objective. ...
In 2022 the CVPR Challenge for Learned Image Compression includes a video track which targets to explore technologies for the compression of HD video sequences. The proposed technologies are evaluated through a subjective test at two operating points: 100 kb/s and 1 Mb/s. This contribution proposes to generate coded videos compliant with the latest standardized video coder, Versatile Video Coding (VVC). The primary objective of this candidate is to assess the recent developments in video coding with respect to this standard to measure the progress made by learning based techniques. To this end, this paper explains how to generate video sequences fulfilling the requirements of this challenge, in a reproducible way, targeting the maximum performance for VVC.
