The use of audio and video modalities for Human Activity Recognition (HAR) is common, given the richness of the data and the availability of pre-trained ML models using a large corpus of labeled training data. However, audio and video sensors also lead to significant consumer privacy concerns. Researchers have thus explored alternate modalities that are less privacy-invasive such as mmWave doppler radars, IMUs, motion sensors. However, the key limitation of these approaches is that most of them do not readily generalize across environments and require significant in-situ training data. Recent work has proposed cross-modality transfer learning approaches to alleviate the lack of trained labeled data with some success. In this paper, we generalize this concept to create a novel system called VAX (Video/Audio to 'X'), where training labels acquired from existing Video/Audio ML models are used to train ML models for a wide range of 'X' privacy-sensitive sensors. Notably, in VAX, once the ML models for the privacy-sensitive sensors are trained, with little to no user involvement, the Audio/Video sensors can be removed altogether to protect the user's privacy better. We built and deployed VAX in ten participants' homes while they performed 17 common activities of daily living. Our evaluation results show that after training, VAX can use its onboard camera and microphone to detect approximately 15 out of 17 activities with an average accuracy of 90%. For these activities that can be detected using a camera and a microphone, VAX trains a per-home model for the privacy-preserving sensors. These models (average accuracy = 84%) require no in-situ user input. In addition, when VAX is augmented with just one labeled instance for the activities not detected by the VAX A/V pipeline (~2 out of 17), it can detect all 17 activities with an average accuracy of 84%. Our results show that VAX is significantly better than a baseline supervised-learning approach of using one labeled instance per activity in each home (average accuracy of 79%) since VAX reduces the user burden of providing activity labels by 8x (~2 labels vs. 17 labels).
Robust Activity Recognition for Adaptive Worker-Robot Interaction using Transfer Learning
Human activity recognition (HAR) using machine learning has shown tremendous promise in detecting construction workers’ activities. HAR has many applications in human-robot interaction research to enable robots’ understanding of human counterparts’ activities. However, many existing HAR approaches lack robustness, generalizability, and adaptability. This paper proposes a transfer learning methodology for activity recognition of construction workers that requires orders of magnitude less data and compute time for comparable or better classification accuracy. The developed algorithm transfers features from a model pre-trained by the original authors and fine-tunes them for the downstream task of activity recognition in construction. The model was pre-trained on Kinetics-400, a large-scale video-based human activity recognition dataset with 400 distinct classes. The model was fine-tuned and tested using videos captured from manual material handling (MMH) activities found on YouTube. Results indicate that the fine-tuned model can recognize distinct MMH tasks in a robust and adaptive manner which is crucial for the widespread deployment of collaborative robots in construction.
more »
« less
- Award ID(s):
- 2047138
- NSF-PAR ID:
- 10447842
- Editor(s):
- Turkan, Yelda; Louis, Joseph; Leite, Fernanda; Ergan, Semiha
- Date Published:
- Journal Name:
- 2023 ASCE International Conference on Computing in Civil Engineering
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
more » « less
-
In safety-critical environments, robots need to reliably recognize human activity to be effective and trust-worthy partners. Since most human activity recognition (HAR) approaches rely on unimodal sensor data (e.g. motion capture or wearable sensors), it is unclear how the relationship between the sensor modality and motion granularity (e.g. gross or fine) of the activities impacts classification accuracy. To our knowledge, we are the first to investigate the efficacy of using motion capture as compared to wearable sensor data for recognizing human motion in manufacturing settings. We introduce the UCSD-MIT Human Motion dataset, composed of two assembly tasks that entail either gross or fine-grained motion. For both tasks, we compared the accuracy of a Vicon motion capture system to a Myo armband using three widely used HAR algorithms. We found that motion capture yielded higher accuracy than the wearable sensor for gross motion recognition (up to 36.95%), while the wearable sensor yielded higher accuracy for fine-grained motion (up to 28.06%). These results suggest that these sensor modalities are complementary, and that robots may benefit from systems that utilize multiple modalities to simultaneously, but independently, detect gross and fine-grained motion. Our findings will help guide researchers in numerous fields of robotics including learning from demonstration and grasping to effectively choose sensor modalities that are most suitable for their applications.more » « less
-
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.more » « less
-
The success and impact of activity recognition algorithms largely depends on the availability of the labeled training samples and adaptability of activity recognition models across various domains. In a new environment, the pre-trained activity recognition models face challenges in presence of sensing bias- ness, device heterogeneities, and inherent variabilities in human behaviors and activities. Activity Recognition (AR) system built in one environment does not scale well in another environment, if it has to learn new activities and the annotated activity samples are scarce. Indeed building a new activity recognition model and training the model with large annotated samples often help overcome this challenging problem. However, collecting annotated samples is cost-sensitive and learning activity model at wild is computationally expensive. In this work, we propose an activity recognition framework, UnTran that utilizes source domains' pre-trained autoencoder enabled activity model that transfers two layers of this network to generate a common feature space for both source and target domain activities. We postulate a hybrid AR framework that helps fuse the decisions from a trained model in source domain and two activity models (raw and deep-feature based activity model) in target domain reducing the demand of annotated activity samples to help recognize unseen activities. We evaluated our framework with three real-world data traces consisting of 41 users and 26 activities in total. Our proposed UnTran AR framework achieves ≈ 75% F1 score in recognizing unseen new activities using only 10% labeled activity data in the target domain. UnTran attains ≈ 98% F1 score while recognizing seen activities in presence of only 2-3% of labeled activity samples.more » « less
-
Human activity recognition (HAR) from wearable sensor data has recently gained widespread adoption in a number of fields. However, recognizing complex human activities, postural and rhythmic body movements (e.g., dance, sports) is challenging due to the lack of domain-specific labeling information, the perpetual variability in human movement kinematics profiles due to age, sex, dexterity and the level of professional training. In this paper, we propose a deep activity recognition model to work with limited labeled data, both for simple and complex human activities. To mitigate the intra- and inter-user spatio-temporal variability of movements, we posit novel data augmentation and domain normalization techniques. We depict a semi-supervised technique that learns noise and transformation invariant feature representation from sparsely labeled data to accommodate intra-personal and inter-user variations of human movement kinematics. We also postulate a transfer learning approach to learn domain invariant feature representations by minimizing the feature distribution distance between the source and target domains. We showcase the improved performance of our proposed framework, AugToAct, using a public HAR dataset. We also design our own data collection, annotation and experimental setup on complex dance activity recognition steps and kinematics movements where we achieved higher performance metrics with limited label data compared to simple activity recognition tasks.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- The DOI is not currently available.
