Stay Informed:
Baskin Engineering COVID-19 Information and Resources
Campus Roadmap to Recovery
Zoom Links: Zoom Help | Teaching with Zoom | Zoom Quick Guide

Perceptual Compression for Video Processing and Storage

Speaker Name: 
Amrita Mazumdar
Speaker Title: 
PhD student
Speaker Organization: 
University of Washington
Start Time: 
Friday, February 22, 2019 - 12:30pm
End Time: 
Friday, February 22, 2019 - 1:15pm
Heiner Litz


Compressed videos constitute 70% of Internet traffic, and video
upload growth rates far outpace compute and storage improvement trends.
Leveraging perceptual cues like saliency, regions where viewers focus their
perceptual attention, can reduce compressed video size while maintaining
perceptual quality, but requires significant changes to video codecs and
ignores the data management of this perceptual information. In this talk, I
present Vignette, a new compression technique and storage manager for
perception-based video compression. Vignette complements off-the-shelf
compression software and hardware codec implementations.  Vignette's
compression technique uses a neural network to predict saliency information
used during transcoding, and its storage manager integrates perceptual
information into the video storage system to support a perceptual
compression feedback loop.Vignette's saliency-based optimizations reduce
storage by up to 95% with minimal quality loss, and \name videos lead to
power savings of 50% on mobile phones during video playback. Our results
demonstrate the benefit of embedding information about the human visual
system into the architecture of video storage systems.


Amrita Mazumdar is a PhD student at the University of Washington,
advised by Luis Ceze and Mark Oskin. Her research interests are at the
intersection of computer architecture and virtual reality, including
high-performance processing and storage for large-scale video systems. Her
past work has focused on designing hardware accelerators for applications
such as VR video processing, near-data similarity search, and low-power
computer vision. She received her MS (2017) from the University of
Washington, and her BS (2014) from Columbia University. Amrita has interned
at Facebook Reality Lab/Oculus Research and IBM, and has received a Google
Anita Borg Memorial Fellowship.