arXiv subjects

Zihe Zhao

Publications and source records attributed to Zihe Zhao.

3 recordsLinked to original sources

arXiv2024

An AI-Driven Multimodal Smart Home Platform for Continuous Monitoring and Assistance in Post-Stroke Motor Impairment

At-home rehabilitation for post-stroke patients presents significant challenges, as continuous, personalized care is often limited outside clinical settings. Moreover, the lack of integrated solutions capable of simultaneously monitoring motor recovery and providing intelligent assistance in home environments hampers rehabilitation outcomes. Here, we present a multimodal smart home platform designed for continuous, at-home rehabilitation of post-stroke patients, integrating wearable sensing, ambient monitoring, and adaptive automation. A plantar pressure insole equipped with a machine learning pipeline classifies users into motor recovery stages with up to 94\% accuracy, enabling quantitative tracking of walking patterns during daily activities. An optional head-mounted eye-tracking module, together with ambient sensors such as cameras and microphones, supports seamless hands-free control of household devices with a 100\% success rate and sub-second response time. These data streams are fused locally via a hierarchical Internet of Things (IoT) architecture, ensuring low latency and data privacy. An embedded large language model (LLM) agent, Auto-Care, continuously interprets multimodal data to provide real-time interventions -- issuing personalized reminders, adjusting environmental conditions, and notifying caregivers. Implemented in a post-stroke context, this integrated smart home platform increased mean user satisfaction from 3.9 $\pm$ 0.8 in conventional home environments to 8.4 $\pm$ 0.6 with the full system ($n=20$). Beyond stroke, the system offers a scalable, patient-centered framework with potential for long-term use in broader neurorehabilitation and aging-in-place applications.

cs.HC↗

arXiv2024

Local deployment of large-scale music AI models on commodity hardware

We present the MIDInfinite, a web application capable of generating symbolic music using a large-scale generative AI model locally on commodity hardware. Creating this demo involved porting the Anticipatory Music Transformer, a large language model (LLM) pre-trained on the Lakh MIDI dataset, to the Machine Learning Compilation (MLC) framework. Once the model is ported, MLC facilitates inference on a variety of runtimes including C++, mobile, and the browser. We envision that MLC has the potential to bridge the gap between the landscape of increasingly capable music AI models and technology more familiar to music software developers. As a proof of concept, we build a web application that allows users to generate endless streams of multi-instrumental MIDI in the browser, either from scratch or conditioned on a prompt. On commodity hardware (an M3 Macbook Pro), our demo can generate 51 notes per second, which is faster than real-time playback for 72.9% of generations, and increases to 86.3% with 2 seconds of upfront buffering.

cs.SD↗

arXiv2024

Semantic Communication via Rate Distortion Perception Bottleneck

With the advancement of Artificial Intelligence (AI) technology, next-generation wireless communication network is facing unprecedented challenge. Semantic communication has become a novel solution to address such challenges, with enhancing the efficiency of bandwidth utilization by transmitting meaningful information and filtering out superfluous data. Unfortunately, recent studies have shown that classical Shannon information theory primarily focuses on the bit-level distortion, which cannot adequately address the perceptual quality issues of data reconstruction at the receiver end. In this work, we consider the impact of semantic-level distortion on semantic communication. We develop an image inference network based on the Information Bottleneck (IB) framework and concurrently establish an image reconstruction network. This network is designed to achieve joint optimization of perception and bit-level distortion, as well as image inference, associated with compressing information. To maintain consistency with the principles of IB for handling high-dimensional data, we employ variational approximation methods to simplify the optimization problem. Finally, we confirm the existence of the rate distortion perception tradeoff within IB framework through experimental analysis conducted on the MNIST dataset.

eess.SP↗