In this paper we present a wearable high rate MIMU (magneticinertial measurement unit) based body tracking system. It is designed using low cost state-of-art hardware and MEMS sensors to reduce errors and improve computational latency. Our system allows for high rate data acquisition and sensor fusion at low power budget. It can be used for range of applications from extreme activity capture and biomechanical analysis to clinical evaluation and ambulatory health monitoring/rehabilitation. The package size of sensing nodes is small, and we use textile wires which make it very flexible. Thus entire system can be easily integrated with body worn suit/pants. Up to 7x nodes can be connected without compromising the maximum sampling frequency (1 KHz), with the possibility to add more nodes using additional bridge stations between nodes. The acquisition rate can be preset from 1 KHz to 100 Hz to suit the application or accuracy requirements. To the best of our knowledge, our inertial motion capture system is the first to offer such high rate output at 1 KHz for multiple nodes. The high rate of inertial data provides intrinsic accuracy to sensor fusion as well as capture high frequency features for clinical diagnostics and biomechanical analysis in ambient settings. The system also runs an embedded sensor fusion algorithm for accurate orientation estimation. We introduce a novel accelerometer and magnetometer measurement correction with adaptive sensor covariance approach in EKF, which makes it robust to both magnetic disturbances and body accelerations. Thus it is well suited for indoor human motion analysis and monitoring highly dynamic motion.

We propose a deep learning based framework that learns data-driven temporal priors to perform 3D human pose estimation from six body worn Magnetic Inertial Measurement units sensors. Our work estimates 3D human pose with associated uncertainty from sparse body worn sensors. We derive and implement a 3D angle representation that eliminates yaw angle (or magnetometer dependence) and show that 3D human pose is still obtained from this reduced representation, but with enhanced uncertainty. We do not use kinematic acceleration as input and show that it improves the generalization to real sensor data from different subjects as well as accuracy. Our framework is based on Bi-directional recurrent autoencoder. A sliding window is used at inference time, instead of full sequence (offline mode). The major contribution of our research is that 3D human pose is predicted from sparse sensors with a well calibrated uncertainty which is correlated with ambiguity and actual errors. We have demonstrated our results on two real sensor datasets; DIP-IMU and Total capture and have come up with state-of-art accuracy. Our work confirms that the main limitation of sparse sensor based 3D human pose prediction is the lack of temporal priors. Therefore fine-tuning on a small synthetic training set of target domain, improves the accuracy.

Realistic estimation and synthesis of articulated human motion must satisfy anatomical constraints on joint angles. A data-driven approach is used to learn human joint limits from 3D motion capture datasets. We represent joint constraints with a new formulation (s 1 , s 2 , τ) using swing-twist representation in exponential maps form. Our parameterization is applied on Human3.6M dataset to create the lookup-map for each joint. These maps enable us to generate 'synthetic' datasets in entire joint rotation space of a given joint. A set of neural network discriminators is then trained with synthetic datasets to learn valid/invalid joint rotations. The discriminators achieve accuracy of [ 94.4 − 99.4%] for different joints. We validate precision-accuracy trade-off of discriminators and qualitatively evaluate classified poses with an interactive tool. The learned discriminators can be used as 'priors' for human pose estimation and motion synthesis.

Electroencephalogram (EEG) is widely used for the diagnosis of neurological conditions like epilepsy, neurodegenerative illnesses and sleep related disorders. Proper interpretation of EEG recordings requires the expertise of trained neurologists, a resource which is scarce in the developing world. Neurologists spend a significant portion of their time sifting through EEG recordings looking for abnormalities. Most recordings turn out to be completely normal, owing to the low yield of EEG tests. To minimize such wastage of time and effort, automatic algorithms could be used to provide pre-diagnostic screening to separate normal from abnormal EEG. Data driven machine learning offers a way forward however, design and verification of modern machine learning algorithms require properly curated labeled datasets. To avoid bias, deep learning based methods must be trained on large datasets from diverse sources. This work presents a new open-source dataset, named the NMT Scalp EEG Dataset, consisting of 2,417 recordings from unique participants spanning almost 625 h. Each recording is labeled as normal or abnormal by a team of qualified neurologists. Demographic information such as gender and age of the patient are also included. Our dataset focuses on the South Asian population. Several existing state-of-the-art deep learning architectures developed for pre-diagnostic screening of EEG are implemented and evaluated on the NMT, and referenced against baseline performance on the well-known Temple University Hospital EEG Abnormal Corpus. Generalization of deep learning based architectures across the NMT and the reference datasets is also investigated. The NMT dataset is being released to increase the diversity of EEG datasets and to overcome the scarcity of accurately annotated publicly available datasets for EEG research.