Tata Consultancy Services Limited

Andrew Gigie

Tapas Chakravarty

Achanna Anil Kumar

Smriti Rani

Arpan Pal

Arijit Sinharay

Mariswamy Girish Chandra

Sounak Dey

Arun George

Kriti Kumar

Avik Ghose

Arijit Chowdhury

Anwesha Khasnobish

Rajat Kumar Das

Dibyanshu Jaiswal

Raj Kumar Rakshit

Angshul Majumdar

Archan Misra

Kuchibhotla Aditi

Rishab Khawad

Krishna Kanth Rokkam

Riddhi Panse

Saurabh Sahu

This disclosure relates generally to a method and system for multi-modal image super-resolution. Conventional methods for multi-modal image super-resolution are performed using joint image based filtering, deep learning and dictionary based approaches which require large datasets for training. Embodiments of the present disclosure provide a joint optimization based transform learning framework wherein a high-resolution (HR) image of target modality is reconstructed from a HR image of guidance modality and a low-resolution (LR) image of target modality. A set of parameters, transforms, coefficients and weight matrices are learnt jointly from a training data which includes a HR image of guidance modality, a LR image of target modality and a HR image of target modality. The learnt set of parameters are used for reconstructing a HR image of target modality. The disclosed joint optimization transform learning framework is used in remote sensing, environment monitoring and so on.

Method and system for multi-modal image super-resolution

Gesture recognition is a key requirement for Human Computer Interaction (HCl) and multiple modalities are explored in literature. Conventionally, channel taps are estimated using least square based estimation and tap corresponding to finger motion is tracked. These assume that noise component is negligible and can reduce the tracking accuracy for low SNR. Thus, to mitigate the above-mentioned limitation, the system and method of the present disclosure explore the feasibility of using speaker and microphone setup available in most of smart devices and transmit inaudible frequencies (acoustic) for detecting the human finger level gestures accurately. More specifically, System implements the method for millimeter level finger tracking and low power gesture detection on this tracked gesture. The system uses a subspace based high resolution technique for delay estimation and use microphone pairs to jointly estimate the multi-coordinates of finger movement.

System and method for acoustic based gesture tracking and recognition using spiking neural network

Conventional gesture detection approaches demand large memory and computation power to run efficiently, thus limiting their use in power and memory constrained edge devices. Present application/disclosure provides a Spiking Neural Network based system which is a robust low power edge compatible ultrasound-based gesture detection system. The system uses a plurality of speakers and microphones that mimics a Multi Input Multi Output (MIMO) setup thus providing requisite diversity to effectively address fading. The system also makes use of distinctive Channel Impulse Response (CIR) estimated by imposing sparsity prior for robust gesture detection. A multi-layer Convolutional Neural Network (CNN) has been trained on these distinctive CIR images and the trained CNN model is converted into an equivalent Spiking Neural Network (SNN) via an ANN (Artificial Neural Network)-to-SNN conversion mechanism. The SNN is further configured to detect/classify gestures performed by user(s).

Acoustic system and method based gesture detection using spiking neural networks

This disclosure relates to selection of optimum channel in twin radars for efficient detection of cardiopulmonary signal rates. State-of-the-art solutions involve use of IQ (In-phase and Quadrature) channel radar which need continuous calibration. Distance of the radar from a subject being monitored affects performance. The present disclosure enables enhanced cardiopulmonary signal rate monitoring using a time domain approach, wherein only the data from signal reflected off the radar is considered. The solution is also time window adaptive. Signal property and radar physics-based methods have been implemented for selecting an optimum channel in twin radars thereby enhancing detection of respiration rate and breath rate.

Method and system for optimum channel selection in twin radar for cardiopulmonary signal monitoring

This disclosure relates generally to methods and systems for real time unobtrusive monitoring of physiological signals of a subject confined to a bed. Output of single channel continuous wave (CW) radars are dependent upon distance of the subject from the radar. Dual channel IQ radars are more accurate but are costly and availability in the market is a constraint. The present disclosure provides a cheap and easily replicable pseudo IQ radar based system. The pseudo IQ radar comprises two CW radars placed at a calibrated distance from each other such that optimum points of one CW radar spatially overlaps null points of the other CW radar and phase imbalance is suppressed. Three such pseudo IQ radars are positioned in a predetermined configuration around the subject being monitored. A Supervised Complex Signal Demodulation (SCSD) method configured to suppress amplitude and DC imbalance is also provided for evaluating the physiological signals.

Real time unobtrusive monitoring of physiological signals

The disclosure herein generally relates to image processing, and, more particularly, to a method and system for impurity detection using multi-modal image processing. This system uses a combination of polarization data, and at least one of a depth data and an RGB image data to perform the impurity material detection. The system uses a graph fusion based approach while processing the captured images to detect presence of the impurity material, and accordingly alert the user.

Method and system for impurity detection using multi-modal imaging

This disclosure relates generally to radar based human activity detection, and, more particularly to, systems and methods from radar based human activity detection and three-dimensional (3D) reconstruction of human gestures using configurable panel radar system. Traditional systems and methods may not provide for a separate capturing of top and bottom parts of the human body. Embodiment of the present disclosure overcome the limitations faced by the traditional systems and methods by identifying a user that performed a gesture; detecting each gesture performed by the identified user; generating, by simulating a set of gesture labels, a sensor data and the generated metadata, a two-dimensional (2D) reference database of different speeds of the detected gestures; computing a displacement and a time of the detected gestures via a pattern matching technique; and reconstructing a video of the identified user performing the detected gestures in 3D.

Systems and methods for three-dimensional (3D) reconstruction of human gestures from radar based measurements

This disclosure relates to non-contact inspection of material for identifying and estimate composition of a material under inspection. Traditionally, material inspection is an invasive process involving contact based approaches. A radar-based approach requires placement of the radar at a specific location, which is a challenge since amplitude of the reflected signal, depends on the distance from the material under inspection. The present disclosure addresses this technical problem by providing a Continuous Wave radar-based approach that is based on absolute slope at extrema points on the reflected signal from the material under inspection.

Non-contact material inspection

Energy remains a critical challenge for continuous sensing: with low-capacity batteries, wearable devices require frequent charging. In contrast, installing sensors in everyday 'smart objects', such as kitchen cabinets, household appliances and office equipment, supports ADL detection via indirect observations on human interaction with such objects, but cannot provide individual-specific insights in multi-tenanted environments. The embodiments herein provide a method and system for energy efficient activity recognition and behavior analysis. Architecture disclosed utilizes a hybrid mode of inexpensive, battery-free sensing of physical activities performed by a subject been monitored during his Activities for Daily Living (ADLs). The sensing combines object interaction sensing with person-specific wearable sensing to recognize individual activities in smart spaces. The method and system disclosed quantifies a probabilistic approach that uses longitudinal observations of user-item interactions, over each individual episode, to compute the anomalous behavior of the subject.

Growing community of inventors

Kolkata, India

Andrew Gigie