Max Robotics — Global Service Robot Directory + US Certification

About

Keshav Thapa is a researcher advancing the frontier of human activity recognition (HAR), a critical technology for pervasive computing, ambient assistive living, robotics, and healthcare monitoring. His work tackles the challenging problem of recognizing complex, concurrent, and interleaved human activities—moving beyond simple, single-action recognition. Thapa’s most cited paper, "A Deep Machine Learning Method for Concurrent and Interleaved Human Activity Recognition" (2020, 39 citations), introduces a novel deep learning framework that addresses the intricacies of overlapping and sequential actions. He further refined probabilistic modeling with the "Log-Viterbi algorithm applied on second-order hidden Markov model for human activity recognition" (2018, 11 citations), enhancing accuracy in dynamic environments. His exploration of deep learning architectures includes "Adapted Long Short-Term Memory (LSTM) for Concurrent Human Activity Recognition" (2021, 9 citations), which adapts LSTM networks to better handle the temporal dependencies of complex activities. With cumulative citations approaching 60, Thapa’s contributions are shaping smarter, more responsive systems for elderly care, assistive technologies, and autonomous robotics, making him a notable voice in the evolution of context-aware computing.