Thomas Plötz

Zikang Leng, Yash Jain, Hyeokhyen Kwon et al. · gatech, microsoft-research

Previous work has demonstrated that virtual accelerometry data, extracted from videos using cross-modality transfer approaches like IMUTube, is beneficial for training complex and effective human activity recognition (HAR) models. Systems like IMUTube were originally designed to cover activities that are based on substantial body (part) movements. Yet, life is complex, and a range of activities of daily living is based on only rather subtle movements, which bears the question to what extent systems like IMUTube are of value also for fine-grained HAR, i.e., When does IMUTube break? In this work we first introduce a measure to quantitatively assess the subtlety of human movements that are underlying activities of interest--the motion subtlety index (MSI)--which captures local pixel movements and pose changes in the vicinity of target virtual sensor locations, and correlate it to the eventual activity recognition accuracy. We then perform a "stress-test" on IMUTube and explore for which activities with underlying subtle movements a cross-modality transfer approach works, and for which not. As such, the work presented in this paper allows us to map out the landscape for IMUTube applications in practical scenarios.

Zikang Leng, Edan Eyal, Yingtian Shi et al.

Engagement, which links to attentional, emotional, and cognitive dimensions, plays an important role in learning. In online and video-based learning environments, learners often need to regulate their own interactions with instructional materials. Measuring and reflecting on engagement can therefore support both learners and adaptive learning systems. In this study, we use wearable and camera-based sensing devices to collect physiological and motion signals, including PPG, ECG, EDA, EEG, IMU, heart rate, temperature, and eye-tracking data, to estimate learner engagement. We conducted a user study with 16 participants in a video-based learning scenario, where participants completed learning tasks and provided repeated in-situ self-reports of engagement through brief probes. We develop and evaluate a system for engagement estimation, compare different sensing modalities, and further analyze the feasibility and effectiveness of multimodal modeling for characterizing learner engagement. Across participant-based cross-validation, our model achieves an MAE of 0.81, 83.75% within-1 accuracy, 73.93% binary accuracy, and 68.45% binary Macro-F1, outperforming sensor-free, statistical, deep temporal, foundation-model, and LLM-based baselines. Our results suggest that fine-grained engagement estimation is feasible but inherently noisy, and that practical systems should prioritize lightweight combinations of behavioral and physiological signals over full multimodal instrumentation. We release the EduGage dataset, including synchronized multimodal sensor signals, probe-aligned momentary engagement labels, video metadata, quizzes, and study materials, to support reproducible research on fine-grained sensor-based engagement modeling in self-guided learning.

Zikang Leng, Hyeokhyen Kwon, Thomas Plötz

In human activity recognition (HAR), the limited availability of annotated data presents a significant challenge. Drawing inspiration from the latest advancements in generative AI, including Large Language Models (LLMs) and motion synthesis models, we believe that generative AI can address this data scarcity by autonomously generating virtual IMU data from text descriptions. Beyond this, we spotlight several promising research pathways that could benefit from generative AI for the community, including the generating benchmark datasets, the development of foundational models specific to HAR, the exploration of hierarchical structures within HAR, breaking down complex activities, and applications in health sensing and activity summarization.

Srivatsa P, Thomas Plötz

There has been a resurgence of applications focused on Human Activity Recognition (HAR) in smart homes, especially in the field of ambient intelligence and assisted living technologies. However, such applications present numerous significant challenges to any automated analysis system operating in the real world, such as variability, sparsity, and noise in sensor measurements. Although state-of-the-art HAR systems have made considerable strides in addressing some of these challenges, they especially suffer from a practical limitation: they require successful pre-segmentation of continuous sensor data streams before automated recognition, i.e., they assume that an oracle is present during deployment, which is capable of identifying time windows of interest across discrete sensor events. To overcome this limitation, we propose a novel graph-guided neural network approach that performs activity recognition by learning explicit co-firing relationships between sensors. We accomplish this by learning a more expressive graph structure representing the sensor network in a smart home, in a data-driven manner. Our approach maps discrete input sensor measurements to a feature space through the application of attention mechanisms and hierarchical pooling of node embeddings. We demonstrate the effectiveness of our proposed approach by conducting several experiments on CASAS datasets, showing that the resulting graph-guided neural network outperforms the state-of-the-art method for HAR in smart homes across multiple datasets and by large margins. These results are promising because they push HAR for smart homes closer to real-world applications.

Shuai Shao, Yu Guan, Bing Zhai et al.

Human activity recognition (HAR) is one of the core research themes in ubiquitous and wearable computing. With the shift to deep learning (DL) based analysis approaches, it has become possible to extract high-level features and perform classification in an end-to-end manner. Despite their promising overall capabilities, DL-based HAR may suffer from overfitting due to the notoriously small, often inadequate, amounts of labeled sample data that are available for typical HAR applications. In response to such challenges, we propose ConvBoost -- a novel, three-layer, structured model architecture and boosting framework for convolutional network based HAR. Our framework generates additional training data from three different perspectives for improved HAR, aiming to alleviate the shortness of labeled training data in the field. Specifically, with the introduction of three conceptual layers--Sampling Layer, Data Augmentation Layer, and Resilient Layer -- we develop three "boosters" -- R-Frame, Mix-up, and C-Drop -- to enrich the per-epoch training data by dense-sampling, synthesizing, and simulating, respectively. These new conceptual layers and boosters, that are universally applicable for any kind of convolutional network, have been designed based on the characteristics of the sensor data and the concept of frame-wise HAR. In our experimental evaluation on three standard benchmarks (Opportunity, PAMAP2, GOTOV) we demonstrate the effectiveness of our ConvBoost framework for HAR applications based on variants of convolutional networks: vanilla CNN, ConvLSTM, and Attention Models. We achieved substantial performance gains for all of them, which suggests that the proposed approach is generic and can serve as a practical solution for boosting the performance of existing ConvNet-based HAR models. This is an open-source project, and the code can be found at https://github.com/sshao2013/ConvBoost

Bing Zhai, Yu Guan, Michael Catt et al.

Sleep is a fundamental physiological process that is essential for sustaining a healthy body and mind. The gold standard for clinical sleep monitoring is polysomnography(PSG), based on which sleep can be categorized into five stages, including wake/rapid eye movement sleep (REM sleep)/Non-REM sleep 1 (N1)/Non-REM sleep 2 (N2)/Non-REM sleep 3 (N3). However, PSG is expensive, burdensome, and not suitable for daily use. For long-term sleep monitoring, ubiquitous sensing may be a solution. Most recently, cardiac and movement sensing has become popular in classifying three-stage sleep, since both modalities can be easily acquired from research-grade or consumer-grade devices (e.g., Apple Watch). However, how best to fuse the data for the greatest accuracy remains an open question. In this work, we comprehensively studied deep learning (DL)-based advanced fusion techniques consisting of three fusion strategies alongside three fusion methods for three-stage sleep classification based on two publicly available datasets. Experimental results demonstrate important evidence that three-stage sleep can be reliably classified by fusing cardiac/movement sensing modalities, which may potentially become a practical tool to conduct large-scale sleep stage assessment studies or long-term self-tracking on sleep. To accelerate the progression of sleep research in the ubiquitous/wearable computing community, we made this project open source, and the code can be found at: https://github.com/bzhai/Ubi-SleepNet.

Zikang Leng, Amitrajit Bhattacharjee, Hrudhai Rajasekhar et al.

One of the primary challenges in the field of human activity recognition (HAR) is the lack of large labeled datasets. This hinders the development of robust and generalizable models. Recently, cross modality transfer approaches have been explored that can alleviate the problem of data scarcity. These approaches convert existing datasets from a source modality, such as video, to a target modality (IMU). With the emergence of generative AI models such as large language models (LLMs) and text-driven motion synthesis models, language has become a promising source data modality as well as shown in proof of concepts such as IMUGPT. In this work, we conduct a large-scale evaluation of language-based cross modality transfer to determine their effectiveness for HAR. Based on this study, we introduce two new extensions for IMUGPT that enhance its use for practical HAR application scenarios: a motion filter capable of filtering out irrelevant motion sequences to ensure the relevance of the generated virtual IMU data, and a set of metrics that measure the diversity of the generated data facilitating the determination of when to stop generating virtual IMU data for both effective and efficient processing. We demonstrate that our diversity metrics can reduce the effort needed for the generation of virtual IMU data by at least 50%, which open up IMUGPT for practical use cases beyond a mere proof of concept.

Zikang Leng, Archith Iyer, Thomas Plötz

Human activity recognition (HAR) is often limited by the scarcity of labeled datasets due to the high cost and complexity of real-world data collection. To mitigate this, recent work has explored generating virtual inertial measurement unit (IMU) data via cross-modality transfer. While video-based and language-based pipelines have each shown promise, they differ in assumptions and computational cost. Moreover, their effectiveness relative to traditional sensor-level data augmentation remains unclear. In this paper, we present a direct comparison between these two virtual IMU generation approaches against classical data augmentation techniques. We construct a large-scale virtual IMU dataset spanning 100 diverse activities from Kinetics-400 and simulate sensor signals at 22 body locations. The three data generation strategies are evaluated on benchmark HAR datasets (UTD-MHAD, PAMAP2, HAD-AW) using four popular models. Results show that virtual IMU data significantly improves performance over real or augmented data alone, particularly under limited-data conditions. We offer practical guidance on choosing data generation strategies and highlight the distinct advantages and disadvantages of each approach.

Zikang Leng, Hyeokhyen Kwon, Thomas Plötz

The development of robust, generalized models in human activity recognition (HAR) has been hindered by the scarcity of large-scale, labeled data sets. Recent work has shown that virtual IMU data extracted from videos using computer vision techniques can lead to substantial performance improvements when training HAR models combined with small portions of real IMU data. Inspired by recent advances in motion synthesis from textual descriptions and connecting Large Language Models (LLMs) to various AI models, we introduce an automated pipeline that first uses ChatGPT to generate diverse textual descriptions of activities. These textual descriptions are then used to generate 3D human motion sequences via a motion synthesis model, T2M-GPT, and later converted to streams of virtual IMU data. We benchmarked our approach on three HAR datasets (RealWorld, PAMAP2, and USC-HAD) and demonstrate that the use of virtual IMU training data generated using our new approach leads to significantly improved HAR model performance compared to only using real IMU data. Our approach contributes to the growing field of cross-modality transfer methods and illustrate how HAR models can be improved through the generation of virtual training data that do not require any manual effort.

Harish Haresamudram, Irfan Essa, Thomas Plötz

The emergence of self-supervised learning in the field of wearables-based human activity recognition (HAR) has opened up opportunities to tackle the most pressing challenges in the field, namely to exploit unlabeled data to derive reliable recognition systems for scenarios where only small amounts of labeled training samples can be collected. As such, self-supervision, i.e., the paradigm of 'pretrain-then-finetune' has the potential to become a strong alternative to the predominant end-to-end training approaches, let alone hand-crafted features for the classic activity recognition chain. Recently a number of contributions have been made that introduced self-supervised learning into the field of HAR, including, Multi-task self-supervision, Masked Reconstruction, CPC, and SimCLR, to name but a few. With the initial success of these methods, the time has come for a systematic inventory and analysis of the potential self-supervised learning has for the field. This paper provides exactly that. We assess the progress of self-supervised HAR research by introducing a framework that performs a multi-faceted exploration of model performance. We organize the framework into three dimensions, each containing three constituent criteria, such that each dimension captures specific aspects of performance, including the robustness to differing source and target conditions, the influence of dataset characteristics, and the feature space characteristics. We utilize this framework to assess seven state-of-the-art self-supervised methods for HAR, leading to the formulation of insights into the properties of these techniques and to establish their value towards learning representations for diverse scenarios.

V. Das Swain, H. Kwon, S. Sargolzaei et al.

A comprehensive understanding of collocation can help understand performance outcomes. For university cohorts, this needs data that describes large groups over a long period. Harnessing user devices to infer this, while tempting, is challenged by privacy concerns, power consumption, and maintenance issues. Alternatively, embedding new sensors in the environment is limited by the expense of covering the entire campus. We investigate the feasibility of leveraging WiFi association logs for this purpose. While these provide coarse approximations of location, these are easily obtainable and depict multiple users on campus over a semester. We explore how these coarse collocations are related to individual performance. Specifically, we inspect the association between individual performance and the collocation behaviors of project group members. We study 163 students (in 54 project groups) over 14 weeks. After describing how we determine collocation with the WiFi logs, we present a study to analyze how collocation within groups relates to a student's final score. We find collocation behaviors show a significant correlation (Pearson's r = 0.24) with performance -- better than both peer feedback or individual behaviors like attendance. Finally, we discuss how repurposing WiFi logs can facilitate applications for domains like mental wellbeing and physical health.

Yu Guan, Thomas Ploetz

Recently, deep learning (DL) methods have been introduced very successfully into human activity recognition (HAR) scenarios in ubiquitous and wearable computing. Especially the prospect of overcoming the need for manual feature design combined with superior classification capabilities render deep neural networks very attractive for real-life HAR application. Even though DL-based approaches now outperform the state-of-the-art in a number of recognitions tasks of the field, yet substantial challenges remain. Most prominently, issues with real-life datasets, typically including imbalanced datasets and problematic data quality, still limit the effectiveness of activity recognition using wearables. In this paper we tackle such challenges through Ensembles of deep Long Short Term Memory (LSTM) networks. We have developed modified training procedures for LSTM networks and combine sets of diverse LSTM learners into classifier collectives. We demonstrate, both formally and empirically, that Ensembles of deep LSTM learners outperform the individual LSTM networks. Through an extensive experimental evaluation on three standard benchmarks (Opportunity, PAMAP2, Skoda) we demonstrate the excellent recognition capabilities of our approach and its potential for real-life applications of human activity recognition.

Sourav Bhattacharya, Petteri Nurmi, Nils Hammerla et al.

We propose a sparse-coding framework for activity recognition in ubiquitous and mobile computing that alleviates two fundamental problems of current supervised learning approaches. (i) It automatically derives a compact, sparse and meaningful feature representation of sensor data that does not rely on prior expert knowledge and generalizes extremely well across domain boundaries. (ii) It exploits unlabeled sample data for bootstrapping effective activity recognizers, i.e., substantially reduces the amount of ground truth annotation required for model estimation. Such unlabeled data is trivial to obtain, e.g., through contemporary smartphones carried by users as they go about their everyday activities. Based on the self-taught learning paradigm we automatically derive an over-complete set of basis vectors from unlabeled data that captures inherent patterns present within activity data. Through projecting raw sensor data onto the feature space defined by such over-complete sets of basis vectors effective feature extraction is pursued. Given these learned feature representations, classification backends are then trained using small amounts of labeled training data. We study the new approach in detail using two datasets which differ in terms of the recognition tasks and sensor modalities. Primarily we focus on transportation mode analysis task, a popular task in mobile-phone based sensing. The sparse-coding framework significantly outperforms the state-of-the-art in supervised learning approaches. Furthermore, we demonstrate the great practical potential of the new approach by successfully evaluating its generalization capabilities across both domain and sensor modalities by considering the popular Opportunity dataset. Our feature learning approach outperforms state-of-the-art approaches to analyzing activities in daily living.