fARSHID sALEMI pARIZI

 
banner.jpg

I am a senior machine learning engineer at OctoAI. I specialize in Generative AI training and model optimization on a host of hardwares.

Previously I was a Ph.D. student at the University of Washington, advised by Shwetak Patel in the Ubiquitous Computing (UbiComp) lab. My research focuses on designing novel input solutions and interaction techniques for virtual and augmented reality. Specifically, I explore the design space of input devices that can be used to enable both high-fidelity immersive experiences, everyday productivity and communication tasks.

 
 

PUBLICATIONS

fingerVibration.PNG

ViFin: Harness Passive Vibration to Continuous Micro Finger Writingwith a Commodity Smartwatch

Wenqiang Chen, Lin Chen, Meiyi Ma, Farshid Salemi Parizi, Patel Shwetak, John Stankovic

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT), 2021

Wearable computing platforms, such as smartwatches and head-mounted mixed reality displays, demand new input devices for high-fidelity interaction. We present AuraRing, a wearable magnetic tracking system designed for tracking fine-grained finger mo…

Wearable computing platforms, such as smartwatches and head-mounted mixed reality displays, demand new input devices for high-fidelity interaction. We present AuraRing, a wearable magnetic tracking system designed for tracking fine-grained finger movement. The hardware consists of a ring with an embedded electromagnetic transmitter coil and a wristband with multiple sensor coils. By measuring the magnetic fields at different points around the wrist, AuraRing estimates the five degree-of-freedom pose of the ring. We develop two different approaches to pose reconstruction—a first-principles iterative approach and a closed-form neural network approach. Notably, AuraRing requires no runtime supervised training, ensuring user and session independence. AuraRing has a resolution of 0.1 mm and a dynamic accuracy of 4.4 mm, as measured through a user evaluation with optical ground truth. The ring is completely self-contained and consumes just2.3 mW of power.

The ability to track handheld controllers in 3D space is critical for interaction with head-mounted displays, such as those used in virtual and augmented reality systems. Today’s systems commonly rely on dedicated infrastructure to track the control…

The ability to track handheld controllers in 3D space is critical for interaction with head-mounted displays, such as those used in virtual and augmented reality systems. Today’s systems commonly rely on dedicated infrastructure to track the controller or only provide inertial-based rotational tracking, which severely limits the user experience. Optical inside-out systems offer mobility but require line-of-sight and bulky tracking rings, which limit the ubiquity of these devices. In this work, we present Aura, an inside-out electromagnetic 6-DoF tracking system for handheld controllers. The tracking system consists of three coils embedded in a headmounted display and a set of orthogonal receiver coils embedded in a handheld controller. We propose a novel closed-form and computationally simple tracking approach to reconstruct position and orientation in real time. Our handheld controller is small enough to fit in a pocket and consumes 45mW of power, allowing it to operate for multiple days on a typical battery. An evaluation study demonstrates that Aura achieves a median tracking error of 5.5mm and 0.8° in 3D space within arm’s reach.

AURARING: Precise Electromagnetic Finger Tracking

Farshid Salemi Parizi*, Eric Whitmire*, Shwetak Patel

(*Co-primary authors)

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT), 2020

Aura: Inside-out Electromagnetic Controller Tracking

Farshid Salemi Parizi*, Eric Whitmire*, Shwetak Patel

(*Co-primary authors)

17th ACM International Conference on Mobile Systems, Applications, and Services (MobiSys), 2019

Neglected toilet overflow can cost thousands of dollars. We propose RoyalFlush: a novel non-invasive overflow system meant to detect such events in their early stages and prevent them from escalating. RoyalFlush uses a floating capacitive sensing te…

Neglected toilet overflow can cost thousands of dollars. We propose RoyalFlush: a novel non-invasive overflow system meant to detect such events in their early stages and prevent them from escalating. RoyalFlush uses a floating capacitive sensing technique that relies on the sizable difference between the dielectric constants of water and air for tracking changes in the water level. Capacitive sensing in this way does not require any hardware inside the toilet bowl.

RoyalFlush: Non-invasive Water Level Monitoring to Prevent Toilet Overflows

Farshid Salemi Parizi, Josh Fromm, Shantanu Deshpande, Shwetak Patel

Proceedings of the ACM 8th International Conference on the Internet of Things, (IoT 2018)

Internet of Things (IoT) applications and platforms are becoming increasingly prevalent. To alleviate recurrent battery replacement and maintenance, we propose a novel battery-free, stick-on capacitive energy harvester that harvests the stray electr…

Internet of Things (IoT) applications and platforms are becoming increasingly prevalent. To alleviate recurrent battery replacement and maintenance, we propose a novel battery-free, stick-on capacitive energy harvester that harvests the stray electric field generated around AC power lines without an ohmic connection to earth ground reference, thereby obviating the need for cumbersome scraping of paint on concrete walls or digging a earth ground plate. Furthermore, our harvester does not require any appliance or load to be operating on the power line and can continuously harvest power after deployment.

CapHarvester: A Stick-on Capacitive Energy Harvesting Using Stray Electric Field

Farshid Salemi Parizi*, Manoj Gulati*, Eric Whitmire, Sidhant Gupta, Shobha Sundar Ram, Amarjeet Singh, Shwetak Patel

(*Co-primary authors)

Proceedings of the ACM on Interactive, Mobile, Wearable and Ubiquitous Technologies (IMWUT), 2018

Knowledge of energy flow in a microwatt-class energy harvesting system is essential to reliable deployment and scheduling of sensing, computation, communication, and actuation tasks. However, existing techniques for monitoring energy flow fail to me…

Knowledge of energy flow in a microwatt-class energy harvesting system is essential to reliable deployment and scheduling of sensing, computation, communication, and actuation tasks. However, existing techniques for monitoring energy flow fail to meet the basic requirements for in-situ realtime monitoring systems by failing to be efficient and failing to perform accurately across a wide dynamic range. UMonitor, makes use of a highly power-optimized “Coulomb counting” implementation to achieve less than 1.7 microampere current draw, 94% efficiency in-situ, and high energy flow measurement accuracy across four orders of magnitude.

UMonitor: In-situ Energy Monitoring with Microwatt Power Consumption

Saman Naderiparizi, Aaron N.Parks, Farshid Salemi Parizi, Joshua R.Smith

IEEE RFID 2016 Best Paper Nominee (Top 3 papers)

 Employment

W-Logo_Purple_Hex.png
fb.png

Octoai, Senior ML Engineer

March 2022 - Current

University Of Washington, Research Assistant

Sep 2016 - June 2022, Advisor: Shwetak Patel

Microsoft Research Intern

June - Sep 2020, Manager: Teddy Seyed

Facebook reality labs, Research Intern

June - Dec 2019, Manager: Wolf Kienzle

NEWS

  • June 2022: Graduate from UW.

  • March 2022: Started at octoai as senior machine learning engineer.

  • june 2020: Started INTERNSHIP at Microsoft Research labs in redmond.

  • Jan & FEB 2020: AuraRing coverage BY UW NEWS, News Atlas, Fierce Electronics, Mobile id world, and KING5.

  • OCT 2019: Auraring got accepted to UBicomp 2020.

  • OCT 2019: Attended ACM UIST Symposium in NEW ORleans.

  • june 2019: Started INTERNSHIP at facebook reality labs in redmond.

  • March 2019: Aura got accepted to mobisys 2019.

  • MARCH 2019: farshid salemi parizi won The UW Reality Lab 2019 Research Proposal Competition. More…

  • OCT 2018: ATTENDED ACM Iot conference in santa barbara.

  • MARCH 2018: Eric whitmire and farshid salemi parizi won The UW Reality Lab 2018 Research Proposal Competition. More…

CONTACT ME