AOMedia Research Symposium 2019

AOMedia kicked off its first Research Symposium on October 21-22, bringing together more than 100 attendees, representatives from member companies and researchers from leading universities. The packed, two-day program of stimulating presentations was held in sunny San Francisco.

image-center

According to a Cisco study, with global internet traffic expected to rise three-fold between 2017-2022, of which 82% will be video, AOMedia is focused on developing new formats that set the standard for video compression performance.

Session Abstracts Symposium Agenda

Introduction & Opening Remarks (01)

Adrian Grange, Google

Slides

Prod AV1 at YouTube (02)

Steve Robertson, YouTube

AV1 in Facebook (03)

Jae Hoon Kim, Facebook

AV1 Encoding at Netflix (04)

Liwei Guo, Netflix

Adaptive Optimal Linear Estimators for Enhanced Motion Compensated Prediction (05)

Kenneth Rose, University of California, Santa Barbara

Slides

What Machines Can Learn from Humans About Lossy Compression (06)

Tsachy Weissman, Stanford University

Slides

A Switchable Region-Based Coding Tool for the AV1 Video Codec (07)

Maggie Zhu, Purdue University

Slides

Incorporating Physical Modeling into Deep Generative Networks for Image and Video Compression (08)

Aswin Sankaranarayanan, Carnegie-Mellon University

Slides

Coding Efficiency Evaluation of AV1 Coding Tools (no video)

Ryan Lei, Intel

Slides

An Overview of New Experimental Coding Tools (09)

Sarah Parker, Google

Slides

Evaluating Video Codecs Through Objective and Subjective Assessments (10)

Fan Zhang, Bristol University

Slides

Speeding up VP9 Intra Encoder with Hierarchical Deep Learning Based Partition Prediction (11)

Somdyuti Paul, University of Texas at Austin

Slides

AV1: Nits, Nitpicks and Shortcomings [Things we should fix for AV2] (12)

Nathan Egge, Mozilla

Slides

Learning-Based AV1 Optimization for VoD and RTC Use Cases (13)

Jinaa Liu, Visionular

Slides

AV1 Image File Format AVIF (14)

Cyril Concolato, Netflix

Slides

Applying Video Coding Tools to WebP Images (15)

Pascal Massimino, Google

Slides

Keynote: Opportunities to use Neural Media Compression, George Toderici, Google (16)

Slides

Deep Learning for Image Compression (17)

Yao Wang, New York University

Slides

Deep Neural Network Based Frame Reconstruction For Optimized Video Coding – An AV2 Approach (18)

Dandan Ding, Hangzhou Normal University

Slides

A Generalized Deep Perceptual Optimizer (19)

Yiannis Andreopoulos, iSize

Slides

Perceptually Optimizing Deep Image Compression (20)

Li-Heng, University of Texas at Austin

Slides

On Perceptual Coding: Quality, Content Features and Complexity (21)

Patrick Le Callet, University of Nantes

Slides

Mode-dependent Data-driven Transforms for AV1 (23)

Antonio Ortega, University of Southern California

Slides

Measuring Video Quality with VMAF: Why You Should Care (24)

Christos Bampis, Netflix

Slides

Motion Based Video Frame Interpolation (25)

Anil Kokaram, Trinity College Dublin

Slides

Real-Time AV1 with SVC support in WebRTC (26)

Alex Gouillard, CoSMo

Slides

AV1 in the MilliCast Real-Time (>200ms) Streaming Platform: The System Level Point of View (27)

Richard Blakely, Milicast

Slides

SVT-AV1 Encoder (28)

Nader Mahdi, Intel

Slides

High-efficiency AV1 Compression Using dAV1d and Eve (29)

Ronald Bultje, Two Orioles

Slides

Overview of FOMS Workshop and Open Issues (30)

Michael Dale, Ellation

Closing Panel Discussion (31)