arXiv subjects

Yuqian Hu

Publications and source records attributed to Yuqian Hu.

6 recordsLinked to original sources

arXiv2026

AM-FM: A Foundation Model for Ambient Intelligence Through WiFi

Ambient intelligence, continuously understanding human presence, activity, and physiology in physical spaces, is fundamental to smart environments, health monitoring, and human-computer interaction. WiFi infrastructure provides a ubiquitous, always-on, privacy-preserving substrate for this capability across billions of IoT devices. Yet this potential remains largely untapped, as wireless sensing has typically relied on task-specific models that require substantial labeled data and limit practical deployment. We present AM-FM, the first foundation model for ambient intelligence and sensing through WiFi. AM-FM is pre-trained on 9.2 million unlabeled Channel State Information (CSI) samples collected over 439 days from 20 commercial device types deployed worldwide, learning general-purpose representations via contrastive learning, masked reconstruction, and physics-informed objectives tailored to wireless signals. Evaluated on public benchmarks spanning nine downstream tasks, AM-FM shows strong cross-task performance with improved data efficiency, demonstrating that foundation models can enable scalable ambient intelligence using existing wireless infrastructure.

cs.LG↗

arXiv2025

CSI-Bench: A Large-Scale In-the-Wild Dataset for Multi-task WiFi Sensing

WiFi sensing has emerged as a compelling contactless modality for human activity monitoring by capturing fine-grained variations in Channel State Information (CSI). Its ability to operate continuously and non-intrusively while preserving user privacy makes it particularly suitable for health monitoring. However, existing WiFi sensing systems struggle to generalize in real-world settings, largely due to datasets collected in controlled environments with homogeneous hardware and fragmented, session-based recordings that fail to reflect continuous daily activity. We present CSI-Bench, a large-scale, in-the-wild benchmark dataset collected using commercial WiFi edge devices across 26 diverse indoor environments with 35 real users. Spanning over 461 hours of effective data, CSI-Bench captures realistic signal variability under natural conditions. It includes task-specific datasets for fall detection, breathing monitoring, localization, and motion source recognition, as well as a co-labeled multitask dataset with joint annotations for user identity, activity, and proximity. To support the development of robust and generalizable models, CSI-Bench provides standardized evaluation splits and baseline results for both single-task and multi-task learning. CSI-Bench offers a foundation for scalable, privacy-preserving WiFi sensing systems in health and broader human-centric applications.

eess.SP↗

arXiv2025

Scaling WiFi Sensing for Ubiquitous Home Monitoring: Lessons from Real-World Deployment on Millions of Devices

WiFi-based home monitoring offers compelling advantages over traditional camera and sensor solutions by leveraging existing wireless infrastructure for contactless, privacy-preserving and through-the-wall detection. This paper presents insights from developing and deploying a WiFi-based human monitoring system across real-world residential environments, addressing the gap between academic research and practical deployment. Through a two-year study involving 280 edge devices across 15 homes in 11 U.S. states, collecting over 4 million motion samples, we identify and address four critical deployment challenges previously underexplored in academic settings: (1) false positives from non-human motion sources (pets, robots) that degrade system reliability, (2) hardware heterogeneity in commercial IoT devices causing inconsistent CSI quality, (3) signal interference in multi-user environments limiting individual tracking capabilities, and (4) computational and bandwidth constraints preventing real-time edge processing. The deployed system integrates a biomechanics-based classifier that reduces non-human false alarms from 63.1\% to 8.4\%, a multi-layer sensing quality metric validating device suitability without environment-specific calibration, proximity-based multi-user detection leveraging distributed IoT devices, and a hybrid edge-cloud architecture with ACF-based compression achieving 99.72\% data reduction. The integrated system achieves 92.61\% human motion detection accuracy across diverse uncontrolled home environments. We have successfully deployed home monitoring technology on millions of WiFi routers nationwide and smart IoT devices (e.g., bulbs, plugs) worldwide, demonstrating its viability for large-scale real-world applications. We share these findings to guide future research toward deployable WiFi sensing systems.

eess.SP↗

arXiv2024

Robust Proximity Detection using On-Device Gait Monitoring

Proximity detection in indoor environments based on WiFi signals has gained significant attention in recent years. Existing works rely on the dynamic signal reflections and their extracted features are dependent on motion strength. To address this issue, we design a robust WiFi-based proximity detector by considering gait monitoring. Specifically, we propose a gait score that accurately evaluates gait presence by leveraging the speed estimated from the autocorrelation function (ACF) of channel state information (CSI). By combining this gait score with a proximity feature, our approach effectively distinguishes different transition patterns, enabling more reliable proximity detection. In addition, to enhance the stability of the detection process, we employ a state machine and extract temporal information, ensuring continuous proximity detection even during subtle movements. Extensive experiments conducted in different environments demonstrate an overall detection rate of 92.5% and a low false alarm rate of 1.12% with a delay of 0.825s.

eess.SP↗

arXiv2024

mmID: High-Resolution mmWave Imaging for Human Identification

Achieving accurate human identification through RF imaging has been a persistent challenge, primarily attributed to the limited aperture size and its consequent impact on imaging resolution. The existing imaging solution enables tasks such as pose estimation, activity recognition, and human tracking based on deep neural networks by estimating skeleton joints. In contrast to estimating joints, this paper proposes to improve imaging resolution by estimating the human figure as a whole using conditional generative adversarial networks (cGAN). In order to reduce training complexity, we use an estimated spatial spectrum using the MUltiple SIgnal Classification (MUSIC) algorithm as input to the cGAN. Our system generates environmentally independent, high-resolution images that can extract unique physical features useful for human identification. We use a simple convolution layers-based classification network to obtain the final identification result. From the experimental results, we show that resolution of the image produced by our trained generator is high enough to enable human identification. Our finding indicates high-resolution accuracy with 5% mean silhouette difference to the Kinect device. Extensive experiments in different environments on multiple testers demonstrate that our system can achieve 93% overall test accuracy in unseen environments for static human target identification.

cs.CV↗

arXiv2022

TAG: Boosting Text-VQA via Text-aware Visual Question-answer Generation

Text-VQA aims at answering questions that require understanding the textual cues in an image. Despite the great progress of existing Text-VQA methods, their performance suffers from insufficient human-labeled question-answer (QA) pairs. However, we observe that, in general, the scene text is not fully exploited in the existing datasets -- only a small portion of the text in each image participates in the annotated QA activities. This results in a huge waste of useful information. To address this deficiency, we develop a new method to generate high-quality and diverse QA pairs by explicitly utilizing the existing rich text available in the scene context of each image. Specifically, we propose, TAG, a text-aware visual question-answer generation architecture that learns to produce meaningful, and accurate QA samples using a multimodal transformer. The architecture exploits underexplored scene text information and enhances scene understanding of Text-VQA models by combining the generated QA pairs with the initial training data. Extensive experimental results on two well-known Text-VQA benchmarks (TextVQA and ST-VQA) demonstrate that our proposed TAG effectively enlarges the training data that helps improve the Text-VQA performance without extra labeling effort. Moreover, our model outperforms state-of-the-art approaches that are pre-trained with extra large-scale data. Code is available at https://github.com/HenryJunW/TAG.

cs.CV↗