cv

PhD Research Intern

Dolby Laboratories

Modeled intrinsic and extrinsic feedback design for vocal training system on glasses.

• Prototyped a smart glasses system for vocal training with ongoing patent filing

AI Engineer Intern

AiFi Inc.

Enhanced the customer-item assoication through characterizing item motion through RFID and customer motion through camera in the autonomous store.

• 95.8% RFID-based item motion classification accuracy
• Coherent feature extraction from camera and RFID data

PhD

Universit of Michigan, Ann Arbor

Embedded Systems

Master

University of Michigan, Ann Arbor

Embedded Systems

Qualcomm Innovation Fellowship

Qualcomm

The Qualcomm Innovation Fellowship (QIF) program is focused on recognizing, rewarding, and mentoring PhD and Masters* students across a broad range of technical research, based on Qualcomm's core values of innovation, execution, and teamwork. The program empowers graduate students to take giant steps toward achieving their research goals.

FloHR: Ubiquitous Heart Rate Measurement using Indirect Floor Vibration Sensing

Jesse R. Codling, Jeffrey D. Shulkin, Yen-Cheng Chang, *Jiale Zhang*, Hugo Latapie, Hae Young Noh, Pei Zhang, Yiwen Dong

The paper presents FloHR, an innovative system for contactless heart rate monitoring that utilizes heartbeat-induced floor vibrations. FloHR contributes a sensitive vibration sensing setup that accurately detects minor vibrations from heartbeats at a distance and an algorithmic framework that distinguishes heartbeats from ambient noise, achieving near-medical accuracy even when sensors are up to 2 meters away from the subject

Poster: Drive-by City Wide Trash Sensing for Neighborhood Sanitation Need

Fernandez, Tomas and Chang, Yen Cheng and Codling, Jesse and Dong, Yiwen and *Zhang, Jiale* and Joe-Wong, Carlee and Noh, Hae Young and Zhang, Pei

We propose a framework for labeling and self-training of in-car video to detect trash on the roads, providing a scalable solution for city-wide trash and air-pollution sensing.

Poster Abstract: Vibration-Based Object Classification with Structural Response of Ambient Music

*Zhang, Jiale* and Pati, Shweta and Codling, Jesse and Bannis, Adeola and Ruiz, Carlos and Noh, Hae Young and Zhang, Pei

The paper presents a novel approach to object classification using vibration-based sensing activated by ambient music. By playing music through a shelf with a sound exciter, the vibration responses vary based on the objects placed on the shelf. The system achieves an accuracy of 98.6% in classifying different objects.

Directly Controlling the Perceived Difficulty of a Shooting Game by the Addition of Fake Enemy Bullets

*Zhang, Jiale*

Adjusting the balance between the player's game skill and the difficulty level is one of the most important factors to improve the player's engagement. This work proposes the concept of user-perceived difficulty and investigates its relationship with the actual game difficulty.

Sniffing Out the City - Vehicular Multimodal Sensing for Environmental and Infrastructure Analysis

Gersey, Julia and Aggarwal, Jatin and *Zhang, Jiale* and Codling, Jesse and Zhang, Pei

Assessing urban infrastructure and environment quality at scale remains a challenge. This work presents a multimodal sensing framework that integrates computer vision-based infrastructure analysis with mobile air quality monitoring to explore urban conditions beyond the camera's field of view. A vehicle-mounted system captures video data for semantic segmentation of roads and buildings while an air intake unit collects temperature, humidity, CO2, TVOC, and AQI levels. A preliminary test drive in Ann Arbor demonstrated expected correlations between CO2 and TVOC spikes in dense urban areas, providing a proof of concept for linking environmental sensing with visual urban analysis. Future work will refine sensor calibration, adaptive sampling strategies, and predictive modeling to improve accuracy and scalability.

Deep-Learning-Enabled Microwave-Induced Thermoacoustic Tomography Based on Sparse Data for Breast Cancer Detection

*J. Zhang*, C. Li, W. Jiang, Z. Wang, L. Zhang and X. Wang

This work proposes a novel deep-learning-enabled microwave-induced thermoacoustic tomography (DL-MITAT) modality to address the sparse data reconstruction problem and applies it in breast cancer detection. The applied deep learning network is a domain transform network called feature projection network (FPNet) + ResU-Net.