arXiv subjects

Zhuangzhuang Dai

Publications and source records attributed to Zhuangzhuang Dai.

14 recordsLinked to original sources

arXiv2026

Privacy-Preserving Industrial Ergonomics: mmWave-Based Automated REBA Scoring and Pose Estimation

Work-related Musculoskeletal Disorders (WMSDs) require continuous ergonomic assessments. While Rapid Entire Body Assessment (REBA) is a gold-standard observation tool, manual monitoring is labor-intensive, and vision-based automation leads to privacy concerns. This paper proposes a novel end-to-end multi-task learning framework for privacy-preserving ergonomic assessment using millimetre-wave (mmWave) radar. A spatio-temporal backbone reconstructs 3D human skeletons, which serves as the biomechanical foundation for a subsequent regression head to generate REBA risk scores. To overcome the sparsity of radar point clouds, we utilise a multi-objective loss function incorporating biomechanical limits and temporal smoothness constraints. Furthermore, we implement an oversampling strategy to address the imbalance of high-risk postures in existing datasets. Experimental results on MMFi dataset demonstrate that our framework achieves a Categorical Accuracy of 77.78% and real-time performance with an inference latency of 5.70 ms. Our method reaches a High-risk REBA MAE of 0.93, which significantly outperforms both direct regression and two-stage pipelines in high-risk scenarios, providing a robust solution for non-invasive industrial ergonomic assessment.

cs.CV↗

arXiv2026

Floor Plan-Agnostic Detection of Gait Speed Drifts Using Ambient Sensors

Gait speed is a vital health indicator for older adults, as changes in gait speed can reflect physiological and functional decline. Ambient sensors offer a promising, privacy-preserving solution for continuous in-home monitoring of gait speed; although it is often limited by methods requiring a home floor plan, which is frequently unfeasible. This paper proposes a novel, floor plan-agnostic method to detect gait speed drifts using only sparse ambient sensors. Our approach identifies informative sensor-to-sensor transitions and analyses fluctuations in their duration. For each sequence a non-parametric statistical test detects changes between a recent period and an initial baseline; and daily test results are aggregated to provide a robust drift detection response. We evaluate our method on a simulated dataset across four different home layouts, showing performance comparable to, and in some cases exceeding, a state-of-the-art baseline that requires floor plan information. This work demonstrates a feasible approach for scalable, cost effective gait drift detection monitoring, providing a foundation for future validation in complex real-world environments.

eess.SP↗

arXiv2026

GazeMoE: Perception of Gaze Target with Mixture-of-Experts

Estimating human gaze target from visible images is a critical task for robots to understand human attention, yet the development of generalizable neural architectures and training paradigms remains challenging. While recent advances in pre-trained vision foundation models offer promising avenues for locating gaze targets, the integration of multi-modal cues -- including eyes, head poses, gestures, and contextual features -- demands adaptive and efficient decoding mechanisms. Inspired by Mixture-of-Experts (MoE) for adaptive domain expertise in large vision-language models, we propose GazeMoE, a novel end-to-end framework that selectively leverages gaze-target-related cues from a frozen foundation model through MoE modules. To address class imbalance in gaze target classification (in-frame vs. out-of-frame) and enhance robustness, GazeMoE incorporates a class-balancing auxiliary loss alongside strategic data augmentations, including region-specific cropping and photometric transformations. Extensive experiments on benchmark datasets demonstrate that our GazeMoE achieves state-of-the-art performance, outperforming existing methods on challenging gaze estimation tasks. The code and pre-trained models are released at https://huggingface.co/zdai257/GazeMoE

cs.CV↗

arXiv2026

Integrating Temporal Context into Streaming Data for Human Activity Recognition in Smart Home

With the global population ageing, it is crucial to enable individuals to live independently and safely in their homes. Using ubiquitous sensors such as Passive InfraRed sensors (PIR) and door sensors is drawing increasing interest for monitoring daily activities and facilitating preventative healthcare interventions for the elderly. Human Activity Recognition (HAR) from passive sensors mostly relies on traditional machine learning and includes data segmentation, feature extraction, and classification. While techniques like Sensor Weighting Mutual Information (SWMI) capture spatial context in a feature vector, effectively leveraging temporal information remains a challenge. We tackle this by clustering activities into morning, afternoon, and night, and encoding them into the feature weighting method calculating distinct mutual information matrices. We further propose to extend the feature vector by incorporating time of day and day of week as cyclical temporal features, as well as adding a feature to track the user's location. The experiments show improved accuracy and F1-score over existing state-of-the-art methods in three out of four real-world datasets, with highest gains in a low-data regime. These results highlight the potential of our approach for developing effective smart home solutions to support ageing in place.

cs.LG↗

arXiv2025

Enhanced Sparse Point Cloud Data Processing for Privacy-aware Human Action Recognition

Human Action Recognition (HAR) plays a crucial role in healthcare, fitness tracking, and ambient assisted living technologies. While traditional vision based HAR systems are effective, they pose privacy concerns. mmWave radar sensors offer a privacy preserving alternative but present challenges due to the sparse and noisy nature of their point cloud data. In the literature, three primary data processing methods: Density-Based Spatial Clustering of Applications with Noise (DBSCAN), the Hungarian Algorithm, and Kalman Filtering have been widely used to improve the quality and continuity of radar data. However, a comprehensive evaluation of these methods, both individually and in combination, remains lacking. This paper addresses that gap by conducting a detailed performance analysis of the three methods using the MiliPoint dataset. We evaluate each method individually, all possible pairwise combinations, and the combination of all three, assessing both recognition accuracy and computational cost. Furthermore, we propose targeted enhancements to the individual methods aimed at improving accuracy. Our results provide crucial insights into the strengths and trade-offs of each method and their integrations, guiding future work on mmWave based HAR systems

cs.CV↗

arXiv2025

GazeTarget360: Towards Gaze Target Estimation in 360-Degree for Robot Perception

Enabling robots to understand human gaze target is a crucial step to allow capabilities in downstream tasks, for example, attention estimation and movement anticipation in real-world human-robot interactions. Prior works have addressed the in-frame target localization problem with data-driven approaches by carefully removing out-of-frame samples. Vision-based gaze estimation methods, such as OpenFace, do not effectively absorb background information in images and cannot predict gaze target in situations where subjects look away from the camera. In this work, we propose a system to address the problem of 360-degree gaze target estimation from an image in generalized visual scenes. The system, named GazeTarget360, integrates conditional inference engines of an eye-contact detector, a pre-trained vision encoder, and a multi-scale-fusion decoder. Cross validation results show that GazeTarget360 can produce accurate and reliable gaze target predictions in unseen scenarios. This makes a first-of-its-kind system to predict gaze targets from realistic camera footage which is highly efficient and deployable. Our source code is made publicly available at: https://github.com/zdai257/DisengageNet.

cs.CV↗

arXiv2025

Multi-Head Adaptive Graph Convolution Network for Sparse Point Cloud-Based Human Activity Recognition

Human activity recognition is increasingly vital for supporting independent living, particularly for the elderly and those in need of assistance. Domestic service robots with monitoring capabilities can enhance safety and provide essential support. Although image-based methods have advanced considerably in the past decade, their adoption remains limited by concerns over privacy and sensitivity to low-light or dark conditions. As an alternative, millimetre-wave (mmWave) radar can produce point cloud data which is privacy-preserving. However, processing the sparse and noisy point clouds remains a long-standing challenge. While graph-based methods and attention mechanisms show promise, they predominantly rely on "fixed" kernels; kernels that are applied uniformly across all neighbourhoods, highlighting the need for adaptive approaches that can dynamically adjust their kernels to the specific geometry of each local neighbourhood in point cloud data. To overcome this limitation, we introduce an adaptive approach within the graph convolutional framework. Instead of a single shared weight function, our Multi-Head Adaptive Kernel (MAK) module generates multiple dynamic kernels, each capturing different aspects of the local feature space. By progressively refining local features while maintaining global spatial context, our method enables convolution kernels to adapt to varying local features. Experimental results on benchmark datasets confirm the effectiveness of our approach, achieving state-of-the-art performance in human activity recognition. Our source code is made publicly available at: https://github.com/Gbouna/MAK-GCN

cs.CV↗

arXiv2025

Action Recognition in Real-World Ambient Assisted Living Environment

The growing ageing population and their preference to maintain independence by living in their own homes require proactive strategies to ensure safety and support. Ambient Assisted Living (AAL) technologies have emerged to facilitate ageing in place by offering continuous monitoring and assistance within the home. Within AAL technologies, action recognition plays a crucial role in interpreting human activities and detecting incidents like falls, mobility decline, or unusual behaviours that may signal worsening health conditions. However, action recognition in practical AAL applications presents challenges, including occlusions, noisy data, and the need for real-time performance. While advancements have been made in accuracy, robustness to noise, and computation efficiency, achieving a balance among them all remains a challenge. To address this challenge, this paper introduces the Robust and Efficient Temporal Convolution network (RE-TCN), which comprises three main elements: Adaptive Temporal Weighting (ATW), Depthwise Separable Convolutions (DSC), and data augmentation techniques. These elements aim to enhance the model's accuracy, robustness against noise and occlusion, and computational efficiency within real-world AAL contexts. RE-TCN outperforms existing models in terms of accuracy, noise and occlusion robustness, and has been validated on four benchmark datasets: NTU RGB+D 60, Northwestern-UCLA, SHREC'17, and DHG-14/28. The code is publicly available at: https://github.com/Gbouna/RE-TCN

cs.CV↗

arXiv2023

SoundCount: Sound Counting from Raw Audio with Dyadic Decomposition Neural Network

In this paper, we study an underexplored, yet important and challenging problem: counting the number of distinct sounds in raw audio characterized by a high degree of polyphonicity. We do so by systematically proposing a novel end-to-end trainable neural network (which we call DyDecNet, consisting of a dyadic decomposition front-end and backbone network), and quantifying the difficulty level of counting depending on sound polyphonicity. The dyadic decomposition front-end progressively decomposes the raw waveform dyadically along the frequency axis to obtain time-frequency representation in multi-stage, coarse-to-fine manner. Each intermediate waveform convolved by a parent filter is further processed by a pair of child filters that evenly split the parent filter's carried frequency response, with the higher-half child filter encoding the detail and lower-half child filter encoding the approximation. We further introduce an energy gain normalization to normalize sound loudness variance and spectrum overlap, and apply it to each intermediate parent waveform before feeding it to the two child filters. To better quantify sound counting difficulty level, we further design three polyphony-aware metrics: polyphony ratio, max polyphony and mean polyphony. We test DyDecNet on various datasets to show its superiority, and we further show dyadic decomposition network can be used as a general front-end to tackle other acoustic tasks.

cs.SD↗

arXiv2023

Detecting Worker Attention Lapses in Human-Robot Interaction: An Eye Tracking and Multimodal Sensing Study

The advent of industrial robotics and autonomous systems endow human-robot collaboration in a massive scale. However, current industrial robots are restrained in co-working with human in close proximity due to inability of interpreting human agents' attention. Human attention study is non-trivial since it involves multiple aspects of the mind: perception, memory, problem solving, and consciousness. Human attention lapses are particularly problematic and potentially catastrophic in industrial workplace, from assembling electronics to operating machines. Attention is indeed complex and cannot be easily measured with single-modality sensors. Eye state, head pose, posture, and manifold environment stimulus could all play a part in attention lapses. To this end, we propose a pipeline to annotate multimodal dataset of human attention tracking, including eye tracking, fixation detection, third-person surveillance camera, and sound. We produce a pilot dataset containing two fully annotated phone assembly sequences in a realistic manufacturing environment. We evaluate existing fatigue and drowsiness prediction methods for attention lapse detection. Experimental results show that human attention lapses in production scenarios are more subtle and imperceptible than well-studied fatigue and drowsiness.

cs.RO↗

arXiv2022

OdomBeyondVision: An Indoor Multi-modal Multi-platform Odometry Dataset Beyond the Visible Spectrum

This paper presents a multimodal indoor odometry dataset, OdomBeyondVision, featuring multiple sensors across the different spectrum and collected with different mobile platforms. Not only does OdomBeyondVision contain the traditional navigation sensors, sensors such as IMUs, mechanical LiDAR, RGBD camera, it also includes several emerging sensors such as the single-chip mmWave radar, LWIR thermal camera and solid-state LiDAR. With the above sensors on UAV, UGV and handheld platforms, we respectively recorded the multimodal odometry data and their movement trajectories in various indoor scenes and different illumination conditions. We release the exemplar radar, radar-inertial and thermal-inertial odometry implementations to demonstrate their results for future works to compare against and improve upon. The full dataset including toolkit and documentation is publicly available at: https://github.com/MAPS-Lab/OdomBeyondVision.

cs.RO↗

arXiv2021

Deep Odometry Systems on Edge with EKF-LoRa Backend for Real-Time Positioning in Adverse Environment

Ubiquitous positioning for pedestrian in adverse environment has served a long standing challenge. Despite dramatic progress made by Deep Learning, multi-sensor deep odometry systems yet pose a high computational cost and suffer from cumulative drifting errors over time. Thanks to the increasing computational power of edge devices, we propose a novel ubiquitous positioning solution by integrating state-of-the-art deep odometry models on edge with an EKF (Extended Kalman Filter)-LoRa backend. We carefully compare and select three sensor modalities, i.e., an Inertial Measurement Unit (IMU), a millimetre-wave (mmWave) radar, and a thermal infrared camera, and realise their deep odometry inference engines which runs in real-time. A pipeline of deploying deep odometry considering accuracy, complexity, and edge platform is proposed. We design a LoRa link for positional data backhaul and projecting aggregated positions of deep odometry into the global frame. We find that a simple EKF based fusion module is sufficient for generic positioning calibration with over 34% accuracy gains against any standalone deep odometry system. Extensive tests in different environments validate the efficiency and efficacy of our proposed positioning system.

cs.RO↗

arXiv2021

DeepAoANet: Learning Angle of Arrival from Software Defined Radios with Deep Neural Networks

Direction finding and positioning systems based on RF signals are significantly impacted by multipath propagation, particularly in indoor environments. Existing algorithms (e.g MUSIC) perform poorly in resolving Angle of Arrival (AoA) in the presence of multipath or when operating in a weak signal regime. We note that digitally sampled RF frontends allow for the easy analysis of signals, and their delayed components. Low-cost Software-Defined Radio (SDR) modules enable Channel State Information (CSI) extraction across a wide spectrum, motivating the design of an enhanced Angle-of-Arrival (AoA) solution. We propose a Deep Learning approach to deriving AoA from a single snapshot of the SDR multichannel data. We compare and contrast deep-learning based angle classification and regression models, to estimate up to two AoAs accurately. We have implemented the inference engines on different platforms to extract AoAs in real-time, demonstrating the computational tractability of our approach. To demonstrate the utility of our approach we have collected IQ (In-phase and Quadrature components) samples from a four-element Universal Linear Array (ULA) in various Light-of-Sight (LOS) and Non-Line-of-Sight (NLOS) environments, and published the dataset. Our proposed method demonstrates excellent reliability in determining number of impinging signals and realized mean absolute AoA errors less than $2^{\circ}$.

eess.SP↗

arXiv2020

Demo Abstract: Indoor Positioning System in Visually-Degraded Environments with Millimetre-Wave Radar and Inertial Sensors

Positional estimation is of great importance in the public safety sector. Emergency responders such as fire fighters, medical rescue teams, and the police will all benefit from a resilient positioning system to deliver safe and effective emergency services. Unfortunately, satellite navigation (e.g., GPS) offers limited coverage in indoor environments. It is also not possible to rely on infrastructure based solutions. To this end, wearable sensor-aided navigation techniques, such as those based on camera and Inertial Measurement Units (IMU), have recently emerged recently as an accurate, infrastructure-free solution. Together with an increase in the computational capabilities of mobile devices, motion estimation can be performed in real-time. In this demonstration, we present a real-time indoor positioning system which fuses millimetre-wave (mmWave) radar and IMU data via deep sensor fusion. We employ mmWave radar rather than an RGB camera as it provides better robustness to visual degradation (e.g., smoke, darkness, etc.) while at the same time requiring lower computational resources to enable runtime computation. We implemented the sensor system on a handheld device and a mobile computer running at 10 FPS to track a user inside an apartment. Good accuracy and resilience were exhibited even in poorly illuminated scenes.

cs.CV↗