Harish Haresamudram

I am a fourth year PhD student with the School of Electrical and Computer Engineering (ECE) at Georgia Institute of Techology, Atlanta. I am advised by Prof. Thomas Ploetz and Prof. Irfan Essa. I received my Master’s degree in May 2019 from Georgia Tech, where my thesis studied the role of representations in human activity recognition using wearables. I was advised by Prof. Thomas Ploetz and Prof. David Anderson.

Previously, I was associated with the Learning and Extraction of Acoustic Patterns Labs, Indian Institute of Science where I worked on speaker verification and spoof detection, advised by Prof. Sriram Ganapathy.

Research

My research broadly involves learning representations for time-series data, with a special focus on developing techniques that require minimal supervision. I develop unsupervised and self-supervised learning algorithms for data from wearable sensors, including accelerometers, gyroscopes and intertial measurement units (IMUs).

Recent News

• Sep, 2022: our papers - Clustering of Human Activities from Wearables by Adopting Nearest Neighbors and A personalized approach for developing a snacking detection system using earbuds in a semi-naturalistic setting have been accepted to ISWC 22, and will be presented at the conference in Atlanta during September 2022.
• Sep, 2022: our paper - Assessing the State of Self-Supervised Human Activity Recognition using Wearables has been published in the Sep 2022 issue of IMWUT! The paper will be presented at Ubicomp 2022 in in Atlanta.
• June, 2021: our paper - Contrastive Predictive Coding for Human Activity Recognition has been published in the June 2021 issue of IMWUT! The paper will be presented at Ubicomp 2021 in September 2021.
• May, 2021: starting my summer internship at Facebook Reality Labs! I will be working data from an EMG-based wearable device.
• November, 2020: our paper - Contrastive Predictive Coding for Human Activity Recognition is currently under review.
• September, 2020: presented our paper - Masked Reconstruction Based Self-Supervision for Human Activity Recognition at ISWC 2020.
• August, 2020: our paper has been accepted to the September issue of IMWUT 2020! : IMUTube: Automatic Extraction of Virtual on-body Accelerometry from Video for Human Activity Recognition. It will be presented at Ubicomp 2020.
• September, 2019: presented our paper - On the Role of Features in Human Activity Recognition at ISWC 2019.

Reviewing

• ISWC 2020, 2021.
• IEEE Pervasive Computing.
• IMWUT 2021.
• KDD 2021.