Video Compression

VideoNerd

Abstract

This document contains results (bitrate/visual quality) of applying x265 (in ABR mode) on two 4K streams (Sintel and TOS), with different lookahead buffer sizes (in frame slots: 3 frames and 20 frames), with two different bitrates 30Mbps and 20Mbps and with two cutree values on/off.

For video quality both  PSNR and VIFP metrics are used. 

Note:  x265 does not allow lookahead buffer size in frames less or equal to the number of consecutive B-frames.

Original streams

All sequences taken from https://media.xiph.org/

Sintel – 3840×1744 resolution, frame rate  24fps.

TOS – resolution 3840×1712, frame rate 24fps

x265 Configuration

All streams are coded with the following fixed parameters:

max CU size = 64×64, min CU size =  8×8

max TU size = 32×32, TU depth = 1 (for both inter and intra)

ME = hex, range =57, subpel =2 , merge =2

Scenecut=0

Gop size = 30 frames

Bframes = 2 (IPBBPBB..),  badapt  = yes, no b-pyramid 

Refs = 2

Signhide = yes

Tmvp = yes

strong-intra-smoothing = yes (since 32×32 TU used)

deblock, sao = yes

Results

Note:  In each table psnr and vifp quality statistics is presented as a pair (mean, std).

Sintel (117 frames, 3840×1744), Cutree = OFF

Lookahead/Expected Bitrate 30 Mbps 20 Mbps
3 frames 37.8 Mbps

Psnr (m,std): 44.574, 3.232

Vifp (m,std): 0.832, 0.055

25.7 Mbps

Psnr (m,std): 43, 3.4

Vifp (m,std): 0.794, 0.06

20 frames 37.8

Psnr (m,std): 44.574, 3.232

Vifp (m,std): 0.832, 0.055

similar results as

in the above cell

TOS (128 frames, 3840×1712), Cutree = OFF

Lookahead/Expected Bitrate 30 Mbps 20 Mbps
3 frames 25.4 Mbps

Psnr (m,std): 42.851, 5.673

Vifp (m,std): 0.711,  0.117

17.6 Mbps

Psnr (m,std): 42.333, 5.600

Vifp (m,std): 0.695, 0.117

20 frames similar results as

in the above cell

similar results as

in the above cell

Sintel (117 frames, 3840×1744),  Cutree = ON

Lookahead/Expected Bitrate 30 Mbps 20 Mbps
3 frames 39.5 Mbps

Psnr: 44.726, 3.175

Vifp: 0.835, 0.057

27 Mbps

Psnr: 43.157, 3.384

Vifp 0.797, 0.068

20 frames 38.8 Mbps

Psnr: 44.842, 3.346

Vifp: 0.840, 0.060

26.8 Mbps

Psnr: 43.385, 3.514

Vifp 0.806, 0.071

TOS (128 frames, 3840×1712), Cutree = ON

Lookahead/Expected Bitrate 30 Mbps 20 Mbps
3 frames 22.8 Mbps

Psnr (m,std): 42.660, 5.530

Vifp (m,std): 0.706, 0.113

15.8 Mbps

Psnr (m,std): 42.180,5.452

Vifp (m,std): 0.691, 0.113

20 frames 22 Mbps

Psnr (m,std): 42.635, 5.509

Vifp (m,std): 0.706, 0.113

15.2 Mbps

Psnr (m,std): 42.183, 5.423

Vifp (m,std): 0.692, 0.111

 

Conclusions

Lookahead buffer length does not impact significantly on coding efficiency regardless target bitrates or cutree values. Therefore minimal lookahead buffer (say 3 frame slots) is sufficient.

15 Responses

  1. I wanted to compose you that bit of observation just to give many thanks yet again relating to the great information you have featured in this case. It has been so open-handed with people like you to supply freely all that numerous people would have offered for sale as an e book to earn some money on their own, even more so considering that you could have done it if you ever considered necessary. The ideas likewise worked like the fantastic way to be aware that most people have the same dream much like my own to find out great deal more when it comes to this matter. I am certain there are a lot more pleasant situations up front for many who check out your site.

  2. Somebody essentially help to make seriously posts I would state. This is the first time I frequented your web page and thus far? I surprised with the research you made to make this particular publish incredible. Fantastic job!

  3. Can I just say what a relief to find someone who actually knows what theyre talking about on the internet. You definitely know how to bring an issue to light and make it important. More people need to read this and understand this side of the story. I cant believe youre not more popular because you definitely have the gift.

  4. Thank you a lot for providing individuals with a very remarkable chance to check tips from here. It is always so useful and stuffed with fun for me personally and my office peers to search your web site really 3 times every week to see the new items you will have. And lastly, I’m just certainly pleased with all the brilliant guidelines you give. Some 3 tips in this article are undoubtedly the most impressive we’ve ever had.

  5. Hmm it looks like your website ate my first comment (it was extremely long) so I guess I’ll just sum it up what I submitted and say, I’m thoroughly enjoying your blog. I as well am an aspiring blog blogger but I’m still new to the whole thing. Do you have any points for inexperienced blog writers? I’d definitely appreciate it.

  6. I’ve been exploring for a bit for any high quality articles or blog posts on this sort of area . Exploring in Yahoo I at last stumbled upon this web site. Reading this information So i am happy to convey that I have an incredibly good uncanny feeling I discovered exactly what I needed. I most certainly will make sure to do not forget this site and give it a glance on a constant basis.

  7. Hey there just wanted to give you a quick heads up. The text in your post seem to be running off the screen in Firefox. I’m not sure if this is a format issue or something to do with browser compatibility but I thought I’d post to let you know. The layout look great though! Hope you get the problem solved soon. Many thanks

    1. to advertise my site and to make it to appear at the top i need pay money.
      i invested all money to web-design, bear in mind this site is completely non-profitable.

Leave a Reply

Your email address will not be published. Required fields are marked *