Arrow Research search

Author name cluster

MANOJ SHARMA

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

2 papers
1 author row

Possible papers

2

NeurIPS Conference 2025 Conference Paper

CogPhys: Assessing Cognitive Load via Multimodal Remote and Contact-based Physiological Sensing

  • Anirudh Bindiganavale Harish
  • Peikun Guo
  • Bhargav Ghanekar
  • Diya Gupta
  • Akilesh Rajavenkatanarayan
  • MANOJ SHARMA
  • Maureen August
  • Akane Sano

Remote physiological sensing is an evolving area of research. As systems approach clinical precision, there is increasing focus on complex applications such as cognitive state estimation. Hence, there is a need for large datasets that facilitate research into complex downstream tasks such as remote cognitive load estimation. A first-of-its-kind, our paper introduces an open-source multimodal multi-vital sign dataset consisting of concurrent recordings from RGB, NIR (near-infrared), thermal, and RF (radio-frequency) sensors alongside contact-based physiological signals, such as pulse oximeter and chest bands, providing a benchmark for cognitive state assessment. By adopting a multimodal approach to remote health sensing, our dataset and its associated hardware system excel at modeling the complexities of cognitive load. Here, cognitive load is defined as the mental effort exerted during tasks such as reading, memorizing, and solving math problems. By using the NASA-TLX survey, we set personalized thresholds for defining high/low cognitive levels, enabling a more reliable benchmark. Our benchmarking scheme bridges the gap between existing remote sensing strategies and cognitive load estimation techniques by using vital signs (such as photoplethysmography (PPG) and respiratory waveforms) and physiological signals (blink waveforms) as an intermediary. Through this paper, we focus on replacing the need for intrusive contact-based physiological measurements with more user-friendly remote sensors. Our benchmarking demonstrates that multimodal fusion significantly improves remote vital sign estimation, with our fusion model achieving $<3~BPM$ (beats per minute) error for vital sign estimation. For cognitive load classification, the combination of remote PPG, remote respiratory signals, and blink markers achieves $86. 49$% accuracy, approaching the performance of contact-based sensing ($87. 5$%) and validating the feasibility of non-intrusive cognitive monitoring.

PDF Details

EAAI Journal 2025 Journal Article

Simple yet robust markerless motion capture system using deep learning

  • Avinash Upadhyay
  • Ankit Shukla
  • MANOJ SHARMA
  • Ajith Abraham

Absolute Three Dimensional human pose estimation (3DHPE) is essential for a wide range of applications, including virtual reality, animation, sports analytics, and biomechanical studies. Traditional marker-based motion capture systems, however, remain expensive, complex to set up, and dependent on specialized hardware, limiting their broader adoption. In this work, we present a unified and accessible marker-less motion capture system designed to overcome these limitations by leveraging a multi-view camera setup composed of low-cost Internet of Things (IoT) devices, such as Raspberry Pi units and a deep learning algorithm. Our pipeline begins by extracting two dimension (2D) poses from each camera view, which are then fused using a learned triangulation method to generate an initial set of absolute three dimension (3D) pose points. To further enhance accuracy, these poses are processed by a spatio-temporal transformer that models the dynamics and relationships among pose points across consecutive frames, ensuring both spatial precision and temporal coherence. The network is trained with additional constraints on bone length consistency and absolute pose error to improve robustness in real-world scenarios. For experimentation, the system is deployed on a local server to demonstrate real-time performance and ease of deployment. We validate our system on the Human3. 6 m and Panoptic datasets, achieving state-of-the-art results while significantly reducing the cost and complexity of 3D human motion capture. This work opens up new possibilities for democratizing high-quality motion capture in both research and industry settings.

Details DOI
v2026.09.13