Mon. Dec 9th, 2019

A smart ring for AR, real-time AI for healthcare, and more University of Washington tech innovations

7 min read
Farshid Salemi Parizi, a PhD student in computer science at the UW, walks through a demo of his team’s AuraRing project: “Precise Electromagnetic Finger Tracking via Smart Ring.” (GeekWire Photos / Taylor Soper)

How can we help visually-impaired software developers edit webpage designs? What technology enables in-home sound awareness for people who are hard of hearing? How will artificial intelligence help healthcare professionals in real-time? Will a high-tech ring improve interaction with an augmented reality world?

These are just a sampling of questions that University of Washington students are trying to answer — and their solutions could very well turn into game-changing innovations that impact people across the world.

The Paul G. Allen School held its annual research showcase Wednesday evening in the new Bill & Melinda Gates Center for Computer Science & Engineering , where alumni, affiliates, investors, entrepreneurs, and others from the community had a chance to learn about the latest work going on at the Seattle campus.

This year’s showcase was held at the new Bill & Melinda Gates Center for Computer Science & Engineering that opened earlier this year and expanded the capacity of the UW’s computer science school.

The event is somewhat akin to a science fair — just one that features some of the country’s top students and professors, working together on a wide-range of tech-related projects.

“It’s amazing to walk around the building and see how much unbelievable work is going on here, the enthusiasm that all the students have for the work they are doing, and the impact they want to have on the world,” said Hank Levy, the former longtime Allen School director who stepped down in July but remains involved with the program.

Ed Lazowska, Bill & Melinda Gates Chair in Computer Science & Engineering, said this year’s batch of projects reflected a focus on solving societal challenges — something he says distinguishes UW from other schools.

“These projects make individual lives better — not just by giving them technology, but by helping them exist in the world in a better way,” he told GeekWire.

Case in point: Venkatesh Potluri, a second-year PhD student, is part of a team building tools to help blind and visually impaired developers create visual content online.

“The really big goal I have is, how can we enable people who are blind to be productive software engineers?” Potluri said.

Jennifer Brennan, a third-year PhD student in computer science, talks about her project: “Estimating the Number and Effect Sizes of Non-null Hypotheses.”

Another theme this year: the amount of cross-disciplinary work going on at the UW campus. The projects featured groups made up of not only students studying computer science, but those in other departments such as biochemistry and electrical engineering.

“It’s really exciting,” said Ishan Chatterjee, a first-year grad student who is helping develop finger tracking via a smart ring. “Academia is unique in that there’s a lot of very smart, very creative people tackling a wide variety of problems.”

Katie Doroschak, a PhD student in computer science, was on a team from the UW’s Molecular Information Systems Lab (MISL) that works on DNA-related technology.

“It’s super fun to work in this environment — I learn so much,” she said. “I don’t necessarily know the intricacies of the biology, but I still get to have exposure.”

Dhruv Jain (right), a second-year PhD student in computer science, joins UW professor Ed Lazowska on stage after his team won an award at the research showcase Thursday.

Several folks from Madrona Venture Group were in attendance Thursday to help hand out the 14th annual Madrona Prize . The Seattle-based venture capital firm has backed 18 startups that have spun out of the UW — most recently OctoML , which started as a research project at the Allen School. Others include Impinj, Skytap, and Turi.

Tim Porter, Madrona managing director, said the event is consistently one of the firm’s favorite nights of the year.

“It’s so inspiring to see all the unbelievable research,” he said. “What a special time it’s been to see the massive impact that the computer science department has had on the overall region, and the stature of the Pacific Northwest and Seattle in particular within the global technology ecosystem.”

UW students Joseph Janizek (left) and Gabriel Erion (center) with Madrona Managing Director Tim Porter. Janizek and Erion won the top Madrona Prize for their project CoAI, which aims to help healthcare professionals in real-time with artificial intelligence.

Check out the winners of the Madrona Prize below, along with all of the projects.

  • Madrona Prize winner: CoAI: Cost Aware Artificial Intelligence for Health Care
  • Madrona Prize runner-up: AuraRing: Precise Electromagnetic Finger Tracking via Smart Ring
  • Madrona Prize runner- up: HomeSound: Exploring Sound Awareness In The Home For People Who Are Deaf And Hard Of Hearing
  • Madrona Prize runner-up: Molecular tagging with nanopore-orthogonal DNA strands
  • People’s Choice winner: HomeSound: Exploring Sound Awareness In The Home For People Who Are Deaf And Hard Of Hearing
  • People’s Choice winner: ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks

All projects:

UbiComp + HCI & Makeability 

  • AuraRing: Precise Electromagnetic Finger Tracking via Smart Ring (Farshid Salemi Parizi, Eric Whitmire, Alvin Cao, Tianke Li, Ishan Chatterjee, Shwetak Patel)
  • Characterizing the Mobile Microtasking Experience (Tal August, Shamsi Iqbal, Michael Gamon, Mark Encarnación)
  • Digital Fabrication of Mechanical and Electrical Functions (Liang He, Jon Froehlich
  • Doppler Exercise Sensing: Smartphone Doppler Ultrasound for Improved Activity Quantification (Parker Ruth, Abhinav Bandari, Anshita Saini, Libby Lavitt, Cindy Lin, Sara Mosiman, Samuel Browd, and Shwetak Patel
  • Heterogeneous Bitwidth Binary Networks (Josh Fromm, Matthai Philipose, Shwetak Patel)
  • HomeSound: Exploring sound awareness in the home for people who are deaf and hard of hearing (Dhruv Jain, Kelly Mack, Steven Goodman, Leah Findlater, Jon Froehlich)
  • In-Home Tracking and Liquid Level Detection of Household Items (Farshid Salemi Parizi, Hanchuan Li, Alvin Cao, Shwetak Patel)
  • Leveraging Routine Behavior and Contextually-Filtered Features for Depression Detection among College Students (Xuhai (Orson) Xu, Prerna Chikersal, Afsaneh Doryab, Daniella Villalba, Janine Dutcher, Michael Tumminia, Tim Althoff, Sheldon Cohen, Kasey Creswell, David Creswell, Jennifer Mankoff, Anind K. Dey)
  • A Multi-Modal Approach for Blind and Visually Impaired Developers to Edit Webpage Designs (Venkatesh Potluri, Liang He, Christine Chen, Jon E. Froehlich, Jennifer Mankoff)
  • O-pH: Optical pH monitor to measure and locate acidity of oral plaque and assist in prediction of tooth decay (Manuja Sharma, Lauren Lee, Matthew Carson, Len Nelson, Shwetak Patel, Eric J Seibel)
  • OsteoApp – Towards mobile osteoporosis screening (Parker Ruth, Edward Jay Wang, Shwetak Patel)
  • Passively-sensed Behavioral Correlates of Discrimination Events in College Students (Yasaman S. Sefidgar, Woosuk Seo, Kevin S. Kuehn, Tim Althoff, Anne Browning, Eve Riskin, Paula S. Nurius, Anind K. Dey, Jennifer Mankoff)
  • Redesigning rapid diagnostic tests with electro-magnetic sensing (Varun Perumal, Manuja Sharma, Shwetak Patel)
  • Seismo – Mobile Blood Pressure Monitoring Using Built-In Smartphone Hardware (Edward Wang, Parker Ruth, Junyi Zhu, Mohit Jain, Tien-Jui Lee, Elliot Saba, Lama Nachman, Shwetak Patel)
  • Social App Accessibility for Deaf Signers (Kelly Mack, Danielle Bragg, Meredith Morris, Maarten Bos, Isabelle Albi, Andres Monroy-Hernandez)
  • Supporting Smartphone-Based Image Capture of RapidDiagnostic Tests in Low-Resource Settings (Chunjong Park, Alex Mariakakis, Shwetak Patel)
  • WhoseCough: In-The-Wild Cougher Verification Using Multitask Learning (Matt Whitehill, Jake Garrison, Shwetak Patel)

Computational Biology / Machine Learning / Robotics 

  • Adversarial Training for Robust Classification of Chest Radiographs (Joseph Janizek, Gabriel Erion, Alex Degrave, Su-In Lee)
  • Avocado: Deep Tensor Factorization for Characterizing the Human Epigenome (Jacob Schreiber, Timothy Durham, Jeffrey Bilmes, William Noble)
  • Estimating the number and effect sizes of non-null hypotheses (Jennifer Brennan, Ramya Korlakai Vinayak, Kevin Jamieson)
  • The Best of Both Modes: Separately Leveraging RGB and Depth for Unseen Object Instance Segmentation (Christopher Xie, Yu Xiang, Arsalan Mousavian, Dieter Fox)
  • CoAI: Cost-Aware Artificial Intelligence for Health Care (Gabriel Erion, Joseph Janizek, Carly Hudelson, Nathan White, Su-In Lee)

AR/VR, Graphics, and Vision

  • ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks (Mohit Shridhar, Jesse Thomason, Daniel Gordon, Yonatan Bisk, Winson Han, Roozbeh Mottaghi, Luke Zettlemoyer, Dieter Fox)
  • Background Matting: The World is Your Green Screen (Soumyadip Sengupta, Vivek Jayaram, Brian Curless, Steve Seitz, Ira Kemelmacher)
    Facial Motion Retargeting from 2D images to 3D characters (Bindita Chaudhuri, Noranart Vesdapunt, Alex Colburn, Linda Shapiro, Baoyuan Wang)
  • High-level Knitting (Benjamin Jones, Yuxuan Mei, Haisen Zhao, Taylor Gotfrid, Jennifer Mankoff, Adriana Schulz)
  • Metasurface Optics for Ultra-Compact Augmented Reality (AR) Visors (Elyas Bayati, Shane Colburn, Arka Majumdar)
  • Replaying Lebron (Luyang Zhu, Kostas Rematas, Steve Seitz, Brian Curless, Ira Kemelmacher-Shlizerman)
  • Reverse Engineering for Fabrication (James Noeckel, Adriana Schulz)
  • Scene Recomposition by Learning-based ICP (Hamid Izadinia, Steve Seitz)
  • Secure Multi-User Content Sharing for Augmented Reality Applications (Kimberly Ruth, Tadayoshi Kohno, and Franziska Roesner)
  • Single-shot three-dimensional imaging with a metasurface depth camera (Shane Colburn and Arka Majumdar)
  • Visual Reaction: Learning To Play Catch With Your Drone (Kuo-Hao Zeng, Roozbeh Mottaghi, Luca Weihs, Ali Farhadi)

Molecular Information Systems Lab / Networking & Systems / SAMPL / Networks

  • Bring Your Own Datatypes: Enabling Datatype Exploration in Deep Learning with TVM (Gus Smith, Luis Vega, Tianqi Chen, Thierry Moreau, Luis Ceze)
  • Fine-Grained Replicated State Machines for a Cluster Storage System (Ming Liu, Arvind Krishnamurthy, Harsha V. Madhyastha, Rishi Bhardwaj, Karan Gupta, Chinmay Kamat, Huapeng Yuan, Aditya Jaltade, Roger Liao, Pavan Konka, Anoop Jawahar)
  • Molecular tagging with nanopore-orthogonal DNA strands (Katie Doroschak, Karen Zhang, Melissa Queen, Aishwarya Mandyam, Karin Strauss, Jeff Nivala, and Luis Ceze)
  • Nanopore readout for scalable DNA circuit reporting (Karen Zhang, Yuan-Jyue Chen, Katie Doroschak, Karin Strauss, Luis Ceze, Jeff Nivala)
  • Nexus: A GPU Cluster Engine for Accelerating DNN-Based Video Analysis (Haichen Shen, Lequn Chen, Yuchen Jin, Liangyu Zhao, Bingyu Kong, Matthai Philipose, Arvind Krishnamurthy, Ravi Sundaram)
  • Peptide Pursuit: Characterizing aptamer binding peptides (Aishwarya Mandyam, Jeff Nivala, Kevin Jamieson, Luis Ceze)
  • Practical, Safe Extensibility for Linux Kernel File Systems (Samantha Miller, Kaiyuan Zhang, Danyang Zhuo, Thomas Anderson)
  • TinySDR, A Software-Defined Radio Platform for Internet of Things (Mehrdad Hessar, Ali Najafi, Vikram Iyer, Shyamnath Gollakota)

Leave a Reply

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

Copyright © All rights reserved. | Newsphere by AF themes.