Conference Proceedings
The magic of Human Visual System perception
Presentation Recording
Description
How Video exploits the HVS using psychological malfunctions to achieve the Quality of Reality:
Video is based off of optical illusions of the HVS which is knowledge we can use to exploit the system and improve our perception of Quality. Subjective and Objective evaluation methods are based upon the HVS knowledge we have accumulated.
Video technology itself has a foundation upon the basic flaw of our human eyes blind spot in allowing motion between frames to be perceived where nothing actually exists. Crunchyroll has a historical legacy of customers demanding high quality above and beyond normal OTT streaming services.
I will go over various incidents with our compression methods as case studies correlating the HVS knowledge base. I will question our conventions and methods held as video engineers for the treatment of video and handling of quality perception.
I will discuss an interesting patent we have obtained covering our work at trying to solve this problem through perceptual frame comparison methods for compression.
The talk will interweave various insights taken from fans of our content over time with the HVS magic of perception angle. We have a unique audience who is very passionate about frame by frame video accuracy without any perceptual quality degradation. Our content is like art, truly drawn frame by frame mostly, and this brings to play a whole new level of quality expectations for our content. Anime is hard to encode as it is, let alone when you are encoding a whole set to serve of all resolutions. I will go into the phenomenon of HLS PTS alignment, debate of not using scenecut, affects this actually has on perceptual quality. Showing how we have conflicting interests at times of streaming smoothly vs. actual encoding quality of each frame to frame in an accurate manner correlating to the Human Visual System itself as ‘Great Quality’.
Presented at Demuxed 2019 in San Francisco.
Other Proceedings
Here are some other proceedings that you might find interesting.
What Codec Should I Use?
Alan Resnick
Doing Server-Side Ad Insertion on Live Sports for 25.3M Concurrent Users
Ashutosh Agrawal
Is now the time to solve the deepfake threat?
Roderick Hodgson
Super Resolution: The scaler of tomorrow, here today!
Nick Chadwick
The do's and don'ts about Streaming security
Javier Brines Garcia
Modeling the conceptual structure of FFmpeg in JavaScript
Ryan Harvey
The magic of Human Visual System perception
Christopher Kennedy
Objectionable Uses of Objective Quality Metrics
Richard Fliam
RTMP: web video innovation or Web 1.0 hack… how did we get to now?
Sarah Allen
Large-Scale Media Archive Migration to the Cloud
Konstantin Wilms
Related Courses
Below are some courses that might interest you.
What Codec Should I Use?
Alan Resnick
Doing Server-Side Ad Insertion on Live Sports for 25.3M Concurrent Users
Ashutosh Agrawal
Is now the time to solve the deepfake threat?
Roderick Hodgson
Super Resolution: The scaler of tomorrow, here today!
Nick Chadwick
The do's and don'ts about Streaming security
Javier Brines Garcia
Modeling the conceptual structure of FFmpeg in JavaScript
Ryan Harvey
The magic of Human Visual System perception
Christopher Kennedy
Objectionable Uses of Objective Quality Metrics
