Next Event

Date: April 15th, 2020
IEEE MTT-SCV Webinar: Area Efficient High-Perf. Voltage Controlled Oscillators in Nano-Scale CMOS

Register

    

Want to volunteer?

The IEEE SCV CAS chapter is seeking volunteers to help with the organization of technical meetings. Please contact us.

    

SCV-CAS Mailing List

To subscribe or unsubcribe, please visit the IEEE SCV-CAS list.

Perception Based Optimization for Video Encoding Systems (Co-sponsored Event)

Date: March 10th, 2020

Jinaa Liu, Visionular CTO, will discuss the recent industry trends in perception oriented methods for optimizing video encoding systems.

About this Event

Event Sponsored and Organized By:

IEEE SPS Chapter of Santa Clara Valley

Location: AMD 2485 Augustine Dr, Santa Clara, CA 95054 (Google Maps)

Venue Details: Please park in parking structure close to the building on Octavius Drive. First building after passing AMD and the road does a curve to the right. Please walk around AMD building to the Highway 101 side to the visitor entrance.

Registration Link: Here.

Schedule:

6:30pm-7:00pm: Registration, Food, Networking

7:00pm-8:00pm: Talk

8:00pm-8:30pm: Q&A and Networking

Cost:

FREE for IEEE members

Suggested donations to cover food and water (pay at door):

Non-IEEE: $5

Students (non-IEEE): $3

Abstract:

In recent years, online video consumption has been growing at an astonishing speed and has dominated the world’s internet traffic. For many video applications, insufficient network bandwidth is the limiting factor for QoE and the need for better video encoding solutions has been higher than ever before. In this talk, we will discuss the recent industry wide trends that aim to optimize video encoding systems with perception oriented approaches, including machine learning based video quality metrics, content adaptive ABR video streaming solutions, ROI guided encoder optimization, the new royalty free codec AV1, and what we have discovered and achieved at Visionular when combining all of these ingredients to serve our customers, particularly for user generated content.

Bio:

Jinaa (Jiayang) Liu, CTO of Visionular – a tech startup based in Mountain View, CA, providing world class video encoding solutions for h.264, h.265, AV1.

Her work includes deep learning based video processing, encoder optimization, and highly efficient distributed transcoding infrastructure.

She previously led software development teams at Google for cross platform real time communication systems (Duo), WebRTC in Chrome and on mobile platforms, and image/audio understanding related research projects at Google AI.

Jinaa received her MS from Rice University and her BS from Tsinghua University.

  • March 2020
    M T W T F S S
     1
    2345678
    9101112131415
    16171819202122
    23242526272829
    3031