We employ Siamese neural networks for autonomous classification of battlefield ground vehicles from acoustic signatures, demonstrating that pairwise similarity learning yields …
HeteroEdge addresses asymmetry in heterogeneous collaborative autonomous systems, proposing mechanisms that allow devices with varying compute, sensing, and energy capabilities to …
HeteroSys investigates heterogeneous and collaborative sensing in real-world deployments, characterizing the trade-offs between device diversity, communication cost, and task …
CHARLIE is a conversational agent that recommends personalized daily fitness and diet plans, combining user profile information with nutrition and activity knowledge to deliver …
PerMTL is a multi-task learning framework for assessing skilled human performance from sensor data, jointly learning shared representations across related assessment tasks to …
Language-guided smart systems can help to design next-generation human-machine interactive applications. The dense text description is one of the research areas where systems learn …
The primary contribution of this paper is designing and prototyping a real-time edge computing system, RhythmEdge, that is capable of detecting changes in blood volume from facial …
An individual’s cognitive and functional abilities are commonly assessed through physical and mental status examination, observational performance measures, surveys and proxy …
Activity Recognition (AR) in older adults living with Neurocognitive disorders caused by diseases such as Alzheimer’s is still a challenging research problem. The inherent natural …
In recent times, wearable devices have gained immense popularity for IoT applications, especially for sports analytics. Recent works in sports analytics primarily focuses on …