Search for: All records

Deep learning has achieved state-of-the-art video action recognition (VAR) performance by comprehending action-related features from raw video. However, these models often learn to jointly encode auxiliary view (viewpoints and sensor properties) information with primary action features, leading to performance degradation under novel views and security concerns by revealing sensor types and locations. Here, we systematically study these shortcomings of VAR models and develop a novel approach, VIVAR, to learn view-invariant spatiotemporal action features removing view information. In particular, we leverage contrastive learning to separate actions and jointly optimize adversarial loss that aligns view distributions to remove auxiliary view information in the deep embedding space using the unlabeled synchronous multiview (MV) video to learn view-invariant VAR system. We evaluate VIVAR using our in-house large-scale time synchronous MV video dataset containing 10 actions with three angular viewpoints and sensors in diverse environments. VIVAR successfully captures view-invariant action features, improves inter and intra-action clusters’ quality, and outperforms SoTA models consistently with 8% more accuracy. We additionally perform extensive studies with our datasets, model architectures, multiple contrastive learning, and view distribution alignments to provide VIVAR insights. We open-source our code and dataset to facilitate further research in view-invariant systems. 

Recent advancements in deep learning-based wearable human action recognition (wHAR) have improved the capture and classification of complex motions, but adoption remains limited due to the lack of expert annotations and domain discrepancies from user variations. Limited annotations hinder the model's ability to generalize to out-of-distribution samples. While data augmentation can improve generalizability, unsupervised augmentation techniques must be applied carefully to avoid introducing noise. Unsupervised domain adaptation (UDA) addresses domain discrepancies by aligning conditional distributions with labeled target samples, but vanilla pseudo-labeling can lead to error propagation. To address these challenges, we propose μDAR, a novel joint optimization architecture comprised of three functions: (i) consistency regularizer between augmented samples to improve model classification generalizability, (ii) temporal ensemble for robust pseudo-label generation and (iii) conditional distribution alignment to improve domain generalizability. The temporal ensemble works by aggregating predictions from past epochs to smooth out noisy pseudo-label predictions, which are then used in the conditional distribution alignment module to minimize kernel-based class-wise conditional maximum mean discrepancy (kCMMD) between the source and target feature space to learn a domain invariant embedding. The consistency-regularized augmentations ensure that multiple augmentations of the same sample share the same labels; this results in (a) strong generalization with limited source domain samples and (b) consistent pseudo-label generation in target samples. The novel integration of these three modules in μDAR results in a range of ~ 4-12% average macro-F1 score improvement over six state-of-the-art UDA methods in four benchmark wHAR datasets. 

The ubiquitousness of smart and wearable devices with integrated acoustic sensors in modern human lives presents tremendous opportunities for recognizing human activities in our living spaces through ML-driven applications. However, their adoption is often hindered by the requirement of large amounts of labeled data during the model training phase. Integration of contextual metadata has the potential to alleviate this since the nature of these meta-data is often less dynamic (e.g. cleaning dishes, and cooking both can happen in the kitchen context) and can often be annotated in a less tedious manner (a sensor always placed in the kitchen). However, most models do not have good provisions for the integration of such meta-data information. Often, the additional metadata is leveraged in the form of multi-task learning with sub-optimal outcomes. On the other hand, reliably recognizing distinct in-home activities with similar acoustic patterns (e.g. chopping, hammering, knife sharpening) poses another set of challenges. To mitigate these challenges, we first show in our preliminary study that the room acoustics properties such as reverberation, room materials, and background noise leave a discernible fingerprint in the audio samples to recognize the room context and proposed AcouDL as a unified framework to exploit room context information to improve activity recognition performance. Our proposed self-supervision-based approach first learns the context features of the activities by leveraging a large amount of unlabeled data using a contrastive learning mechanism and then incorporates this feature induced with a novel attention mechanism into the activity classification pipeline to improve the activity recognition performance. Extensive evaluation of AcouDL on three datasets containing a wide range of activities shows that such an efficient feature fusion-mechanism enables the incorporation of metadata that helps to better recognition of the activities under challenging classification scenarios with 0.7-3.5% macro F1 score improvement over the baselines. 

We explore the effect of auxiliary labels in improving the classification accuracy of wearable sensor-based human activity recognition (HAR) systems, which are primarily trained with the supervision of the activity labels (e.g. running, walking, jumping). Supplemental meta-data are often available during the data collection process such as body positions of the wearable sensors, subjects' demographic information (e.g. gender, age), and the type of wearable used (e.g. smartphone, smart-watch). This information, while not directly related to the activity classification task, can nonetheless provide auxiliary supervision and has the potential to significantly improve the HAR accuracy by providing extra guidance on how to handle the introduced sample heterogeneity from the change in domains (i.e positions, persons, or sensors), especially in the presence of limited activity labels. However, integrating such meta-data information in the classification pipeline is non-trivial - (i) the complex interaction between the activity and domain label space is hard to capture with a simple multi-task and/or adversarial learning setup, (ii) meta-data and activity labels might not be simultaneously available for all collected samples. To address these issues, we propose a novel framework Conditional Domain Embeddings (CoDEm). From the available unlabeled raw samples and their domain meta-data, we first learn a set of domain embeddings using a contrastive learning methodology to handle inter-domain variability and inter-domain similarity. To classify the activities, CoDEm then learns the label embeddings in a contrastive fashion, conditioned on domain embeddings with a novel attention mechanism, enforcing the model to learn the complex domain-activity relationships. We extensively evaluate CoDEm in three benchmark datasets against a number of multi-task and adversarial learning baselines and achieve state-of-the-art performance in each avenue. 

Activity Recognition (AR) models perform well with a large number of available training instances. However, in the presence of sensor heterogeneity, sensing biasness and variability of human behaviors and activities and unseen activity classes pose key challenges to adopting and scaling these pre-trained activity recognition models in the new environment. These challenging unseen activities recognition problems are addressed by applying transfer learning techniques that leverage a limited number of annotated samples and utilize the inherent structural patterns among activities within and across the source and target domains. This work proposes a novel AR framework that uses the pre-trained deep autoencoder model and generates features from source and target activity samples. Furthermore, this AR frame-work establishes correlations among activities between the source and target domain by exploiting intra- and inter-class knowledge transfer to mitigate the number of labeled samples and recognize unseen activities in the target domain. We validated the efficacy and effectiveness of our AR framework with three real-world data traces (Daily and Sports, Opportunistic, and Wisdm) that contain 41 users and 26 activities in total. Our AR framework achieves performance gains ≈ 5-6% with 111, 18, and 70 activity samples (20 % annotated samples) for Das, Opp, and Wisdm datasets. In addition, our proposed AR framework requires 56, 8, and 35 fewer activity samples (10% fewer annotated examples) for Das, Opp, and Wisdm, respectively, compared to the state-of-the-art Untran model.