公开(公告)号：US11631247B2

公开(公告)日：2023-04-18

申请号：US17197316

申请日：2021-03-10

Applicant: Tata Consultancy Services Limited

Inventor： Jayavardhana Rama Gubbi Lakshminarasimha ,  Akshaya Ramaswamy ,  Balamuralidhar Purushothaman ,  Aparna Kanakatte Gurumurthy ,  Avik Ghose

IPC: G06V20/40 ,  G06K9/62 ,  G06N3/08

Abstract: State of the art techniques in the domain of video analysis have limitations in terms of capability to capture spatio-temporal representation. This limitation in turn affects interpretation of video data. The disclosure herein generally relates to video analysis, and, more particularly, to a method and system for video analysis to capture spatio-temporal representation for video reconstruction and analysis. The method presents different architecture variations using three main deep network components: 2D convolution units, 3D convolution units and long short-term memory (LSTM) units for video reconstruction and analysis. These variations are trained for learning the spatio-temporal representation of the videos in order to generate a pre-trained video analysis module. By understanding the advantages and disadvantages of different architectural configurations, a novel architecture is designed for video reconstruction. Using transfer learning, the video reconstruction pre-trained model is extended to other video applications such as video object segmentation and surgical video tool segmentation.

公开(公告)号：US20230047937A1

公开(公告)日：2023-02-16

申请号：US17552868

申请日：2021-12-16

Applicant: Tata Consultancy Services Limited

Inventor： Jayavardhana Rama GUBBI LAKSHMINARASIMHA ,  Vartika Sengar ,  Vivek Bangalore Sampathkumar ,  Aparna Kanakatte Gurumurthy ,  Murali Poduval ,  Balamuralidhar Purushothaman ,  Karthik Seemakurthy ,  Avik Ghose ,  Srinivasan Jayaraman

IPC: G06T5/00 ,  G06T7/269 ,  G06T5/50

Abstract: The disclosure herein relates to methods and systems for generating an end-to-end de-smoking model for removing smoke present in a video. Conventional data-driven based de-smoking approaches are limited mainly due to lack of suitable training data. Further, the conventional data-driven based de-smoking approaches are not end-to-end for removing the smoke present in the video. The de-smoking model of the present disclosure is trained end-to-end with the use of synthesized smoky video frames that are obtained by source aware smoke synthesis approach. The end-to-end de-smoking model localize and remove the smoke present in the video, using dynamic properties of the smoke. Hence the end-to-end de-smoking model simultaneously identifies the regions affected with the smoke and performs the de-smoking with minimal artifacts. localized smoke removal and color restoration of a real-time video.

公开(公告)号：US10960893B2

公开(公告)日：2021-03-30

申请号：US15878857

申请日：2018-01-24

Applicant: Tata Consultancy Services Limited

Inventor： Arijit Chowdhury ,  Avik Ghose ,  Tapas Chakravarty

IPC: G07C5/00 ,  B60W40/09 ,  G01C21/32 ,  G01C21/26 ,  G01S19/14

Abstract: This disclosure relates generally to driver profiling, and more particularly to system and method for driver profiling corresponding to automobile trip. In one embodiment, the method includes selectively computing, based on GPS data captured during a trip, features such that a first set of features are computed at a computation device and a second set of features are computed at a cloud server. The first and second set of features differs in computational complexity level. Said features include statistical data for attributes computed from the GPS data. The features corresponding to the trip are stored at repository associated with the cloud server. The repository includes previously precomputed features data associated with a set of driver profiles for previously completed trips. A driver profile corresponding to the trip is identified from amongst the set of driver profiles based on comparison of the plurality of features and the previously computed features data.

公开(公告)号：US20200305763A1

公开(公告)日：2020-10-01

申请号：US16827756

申请日：2020-03-24

Applicant: Tata Consultancy Services Limited

Inventor： Vivek CHANDEL ,  Shivam Singhal ,  Avik Ghose

IPC: A61B5/103 ,  A61B5/00 ,  G01L1/16 ,  G01P15/18 ,  G01P15/02 ,  G01C19/00 ,  G01P7/00

Abstract: The present disclosure provides wearable apparatus and method for calculating drift-free plantar pressure parameters for gait monitoring of an individual. Most conventional techniques use different kind of sensors placed in in-sole based wearable apparatus but are costly and not effective in calculating accurate plantar pressure parameters. The disclosed wearable apparatus uses off-the shelf piezoelectric sensors that are widely available in market with less cost. The drift-free plantar pressure parameters are calculated using drift-free static pressure data obtained by numerically integrating acquired dynamic sensor data from the piezoelectric sensors, using a LiTCEM correction mechanism. A 6-DOF Inertial Measurement Unit (IMU sensor) helps in isolating zero-pressure duration indicating when a foot of the individual is in air during a stride, while obtaining the drift-free static pressure data. The disclosed wearable apparatus calculate the drift-free plantar pressure parameters for long duration and facilitates monitoring walking patterns of the individual.

公开(公告)号：US20190056426A1

公开(公告)日：2019-02-21

申请号：US15935731

申请日：2018-03-26

Applicant: TATA CONSULTANCY SERVICES LIMITED

Inventor： Vivek Chandel ,  Kartik Muralidharan ,  Avik Ghose ,  Archan Misra

IPC: G01P21/00 ,  G01P3/44 ,  G01C19/5776

CPC classification number: G01P21/00 ,  G01C19/00 ,  G01C19/5776 ,  G01C25/005 ,  G01P3/44

Abstract: Embodiments of the present disclosure provide systems and methods that establish non-linear components of gyroscope errors which have not been studied or explored earlier. These errors include but are not limited to non-linear dynamic error which is a function of the angular velocity itself. Bias instability has been observed within the same environment of temperature and atmospheric pressure. In other words, the embodiments of the present disclosure analyse and models static bias errors and dynamic non-linear errors in the gyroscope sensor which may be used to model and correct errors accordingly In subsequent measurements. The system provide solution(s) when there is no way of directly estimating temperature for bias correction of gyroscope, by estimating a temperature change by considering indirect measurements from other sensors present onboard the device.

公开(公告)号：US09855012B2

公开(公告)日：2018-01-02

申请号：US15073232

申请日：2016-03-17

Applicant: Tata Consultancy Services Limited

Inventor： Rohan Banerjee ,  Anirban Dutta Choudhury ,  Aniruddha Sinha ,  Avik Ghose ,  Arpan Pal

IPC: G06K9/00 ,  A61B5/00 ,  A61B5/021 ,  G06T5/00 ,  G06T5/20 ,  G06T7/00 ,  A61B5/024

CPC classification number: A61B5/7235 ,  A61B5/0059 ,  A61B5/0077 ,  A61B5/021 ,  A61B5/02116 ,  A61B5/02438 ,  A61B5/14551 ,  A61B5/6898 ,  A61B5/7203 ,  A61B5/7246 ,  A61B5/7264 ,  A61B5/7278 ,  G06T5/002 ,  G06T5/20 ,  G06T7/0012 ,  G06T2207/10016 ,  G06T2207/20182 ,  G06T2207/30104

Abstract: A method and system is provided for noise cleaning of photoplethysmogram signals. The method and system is disclosed for noise cleaning of photoplethysmogram signals for estimating blood pressure of a user; wherein photoplethysmogram signals are extracting from the user; the extracted photoplethysmogram signals are up sampled; the up sampled photoplethysmogram signals are filtered; uneven baseline drift of each cycle is removed from the up sampled and filtered photoplethysmogram signals; outlier cycles of the photoplethysmogram signals are removed and remaining cycles of the photoplethysmogram signals are modeled; and time domain features are extracted from originally extracted and modeled photoplethysmogram signals for estimating blood pressure of the user.

公开(公告)号：US20170265020A1

公开(公告)日：2017-09-14

申请号：US15274449

申请日：2016-09-23

Applicant: Tata Consultancy Services Limited

Inventor： Keshaw DEWANGAN ,  Shahnawaz Alam ,  Arijit Sinharay ,  Avik Ghose ,  Arpan Pal

IPC: H04W4/00

CPC classification number: H04W4/38 ,  H04L67/10 ,  H04L67/42

Abstract: A system and a method for mobile sensing data processing are provided. The method includes, receiving one or more requests from one or more applications installed at a client device to obtain a processed sensing data obtained in response to execution of one or more tasks by the application using a set of sensors. Raw data is extracted from the set of sensors in response to the execution of the tasks. A data stream is configured to include sensor data and a task information associated with the tasks. The client device is connected with the server to transmit the data stream. The server outputs the processed sensing data upon processing the data stream and the task information by using one or more task specific models stored at the server. The processed sensing data is received from the server and provided to the applications.

公开(公告)号：US20170006436A1

公开(公告)日：2017-01-05

申请号：US15200781

申请日：2016-07-01

Applicant: Tata Consultancy Services Limited

Inventor： Nasimuddin Ahmed ,  Avik Ghose ,  Tapas Chakravarty ,  Arpan Pal

IPC: H04W4/04 ,  H04W4/02

CPC classification number: H04W4/33 ,  G01S5/0252 ,  H04W4/021 ,  H04W4/043

Abstract: Method(s) and System(s) for determining location of a user device within a premise are described. The method includes identifying multiple zones with physical boundaries within the premise based on parameters associated with geometry of the premise. The premise includes multiple access points distributed across the multiple zones. Thereafter, the method includes collecting a first set of Received Signal Strength Indicator (RSSI) Data that is representative of strength of signals received from each accessible access point, at different locations within the premise. After collecting the first set, the method includes computing a Variable Path Loss Exponent (VPLE) within each zone for each accessible access point for determining location of the user device based on at least one of the first set of RSSI data, a line of sight condition, a non-line of sight condition and distance between each accessible access point from each location.

Abstract translation: 描述用于确定内部用户设备位置的方法和系统。 该方法包括基于与前提的几何结构相关联的参数在前提内识别具有物理边界的多个区域。 前提包括跨多个区域分布的多个接入点。 此后，该方法包括在前提内的不同位置处收集代表从每个可访问接入点接收的信号的强度的第一组接收信号强度指示符（RSSI）数据。 在收集第一组之后，该方法包括针对每个可访问接入点计算每个区域内的可变路径损耗指数（VPLE），以基于第一组RSSI数据，视线中的至少一个来确定用户设备的位置 条件，非线视线条件和每个位置的每个可访问接入点之间的距离。

公开(公告)号：US20160310084A1

公开(公告)日：2016-10-27

申请号：US15073232

申请日：2016-03-17

Applicant: Tata Consultancy Services Limited

Inventor： Rohan Banerjee ,  Anirban Dutta Choudhury ,  Aniruddha Sinha ,  Avik Ghose ,  Arpan Pal

IPC: A61B5/00 ,  G06T7/00 ,  G06T5/20 ,  A61B5/021 ,  G06T5/00

CPC classification number: A61B5/7235 ,  A61B5/0059 ,  A61B5/0077 ,  A61B5/021 ,  A61B5/02116 ,  A61B5/02438 ,  A61B5/14551 ,  A61B5/6898 ,  A61B5/7203 ,  A61B5/7246 ,  A61B5/7264 ,  A61B5/7278 ,  G06T5/002 ,  G06T5/20 ,  G06T7/0012 ,  G06T2207/10016 ,  G06T2207/20182 ,  G06T2207/30104

Abstract: A method and system is provided for noise cleaning of photoplethysmogram signals. The method and system is disclosed for noise cleaning of photoplethysmogram signals for estimating blood pressure of a user; wherein photoplethysmogram signals are extracting from the user; the extracted photoplethysmogram signals are up sampled; the up sampled photoplethysmogram signals are filtered; uneven baseline drift of each cycle is removed from the up sampled and filtered photoplethysmogram signals; outlier cycles of the photoplethysmogram signals are removed and remaining cycles of the photoplethysmogram signals are modeled; and time domain features are extracted from originally extracted and modeled photoplethysmogram signals for estimating blood pressure of the user.

Abstract translation: 提供了一种用于光电体积描记图信号的噪声清洁的方法和系统。 公开了用于估计使用者的血压的光电容积图信号的噪声清洁的方法和系统; 其中光电体积描图信号从用户提取; 提取的光电容积图信号被上采样; 上采样的photoplethysmogram信号被过滤; 从上采样和滤波的光电容积图中消除每个周期的不均匀基线漂移; 除去光电容积图信号的异常周期，并对光电容积图的信号的剩余周期进行建模; 并且从用于估计用户血压的原始提取和建模的光电容积图信号中提取时域特征。

公开(公告)号：US09324192B2

公开(公告)日：2016-04-26

申请号：US14182833

申请日：2014-02-18

Applicant: Tata Consultancy Services Limited

Inventor： Tapas Chakravarty ,  Avik Ghose ,  Balamuralidhar Purushothaman ,  Arpan Pal ,  Arijit Chowdhary

IPC: G07C5/00 ,  G05B23/02

CPC classification number: G07C5/00 ,  G05B23/0235

Abstract: A monitoring unit for vehicle monitoring comprising a receiving module configured to receive data from an OBD, wherein the data is associated with a plurality of jerks detected by a 3-axis accelerometer. The monitoring unit comprises an analytics module configured to compare an intensity of each jerk of the plurality of jerks to a predefined jerk threshold and capture high intensity jerks from the plurality of jerks. The high intensity jerks have intensity equal to or more than the predefined jerk threshold. The method further comprises determining an elapsed time for each of the high intensity jerks. The elapsed time for each of the high intensity jerks is compared to a predefined time threshold. Further it is determined whether an analysis on the high intensity jerks is to be performed at the vehicle or at a server located remotely.

Abstract translation: 一种用于车辆监视的监控单元，包括被配置为从OBD接收数据的接收模块，其中所述数据与由三轴加速度计检测到的多个跳线相关联。 监测单元包括分析模块，其被配置为将多个抖动中的每个冲击的强度与预定的加加速度阈值进行比较，并捕获来自多个跳动的高强度跳动。 高强度强度具有等于或大于预定的加加速度阈值的强度。 该方法还包括确定每个高强度跳跃的经过时间。 将每个高强度抖动的经过时间与预定义的时间阈值进行比较。 此外，确定是否在车辆或位于远程的服务器上执行对高强度刺激的分析。