Netradyne, Inc.

Samsung Electronics Co., Ltd.

Arvind Yedla

David Jonathan Julian

Venkata Sreekanta Reddy Annapureddy

Michael Campos

Mostafa El-Khamy

Veeresh Taranalli

Jungwon Lee

Venkata Ramanan Venkatachalam Jayaraman

Jonathan Albert Cox

Marcel Nassar

Inyup Kang

Sandeep Pandya

Adam David Kahn

Avneesh Agrawal

Badugu Naveen Chakravarthy

Suresh Babu Yanamala

Hyunsang Cho

Jaeyoon Kim

Dale Alan Willis

Sang-Hyuck Ha

Vinay Kumar Rai

Shravan Kumar Motuti

Suresh Kumar Yerakaraju

SangHyuck Ha

Biswarup Debnath

Praveen Mathad

Hitesh Gupta

Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle. Combinations of co-occurring driving events may be detected and used to warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Combination alerts

Systems, devices and methods provide, implement, and use vision-based methods of sequence inference for a device affixed to a vehicle.

Multiple exposure event determination

A system or method trains and implements a machine learning model for object detection. A first computing device remote to a vehicle trains a machine learning model by receiving a decoded set of previously encoded training image data and training the machine learning model to detect objects within images based on the decoded set of training image data. A processor of a second computing device encodes at least one image captured by a camera, decodes the encoded image, and executes the machine learning model to extract a set of objects from the decoded image.

Real-time object detection from decompressed images

Disclosed herein are systems and methods for classifying traffic indicators to detect intersection crossings by a vehicle. A computing device can receive a sequence of frames captured by a capture device mounted to the vehicle, and generate an intersection status data structure for each frame in the sequence of frames using a full-frame classification model. The computing device can also classify a set of features detected in each frame using an object detection model, and detect an intersection crossing event based on the intersection status data structure of each frame and the classification of each of the set of features detected in each frame. The systems and methods described herein improve upon existing traffic light classification models by using a full-frame intersection classification model in connection with traditional object tracking techniques, thereby reducing false-positive crossing-event detection and improving the accuracy of classification in a variety of environmental conditions.

Systems and methods for detecting intersection crossing events using full frame classification techniques

Systems and methods are provided for intelligent driving monitoring systems, advanced driver assistance systems and autonomous driving systems, and providing alerts to the driver of a vehicle, based on anomalies detected between driver behavior and environment captured by the outward facing camera. Various aspects of the driver, which may include his direction of sight, point of focus, posture, gaze, is determined by image processing of the upper visible body of the driver, by a driver facing camera in the vehicle. Other aspects of environment around the vehicle captured by the multitude of cameras in the vehicle are used to correlate driver behavior and actions with what is happening outside to detect and warn on anomalies, prevent accidents, provide feedback to the driver, and in general provide a safer driver experience.

Inward/outward vehicle monitoring for remote reporting and in-cab warning enhancements

The present disclosure provides devices, systems and methods for mapping of traffic boundaries.

Traffic boundary mapping

Apparatuses (including user equipment (UE) and modern chips for UEs), systems, and methods for UE downlink Hybrid Automatic Repeat reQuest (HARQ) buffer memory management are described. In one method, the entire UE DL HARQ buffer memory space is pre-partitioned according to the number and capacities of the UE's active carrier components. In another method, the UE DL HARQ buffer is split between on-chip and off-chip memory so that each partition and sub-partition is allocated between the on-chip and off-chip memories in accordance with an optimum ratio.

Method and system for contiguous HARQ memory management with memory splitting

A neural network is configured to process image data captured by a vehicle-mounted camera. The neural network includes common processing layers (a trunk) and separate, parallelizable processing layers (branches). An object detection branch of the neural network is trained to detect objects that may be visible from the vehicle-mounted camera, such as cars, trucks, and traffic signs. A curve determination branch is trained to detect and parameterize salient curves, such as lane boundaries and road boundaries. The curve determination branch itself is configured with a trunk and branch architecture, having both common and separate processing layers. A first branch computes a likelihood that a curve is present in a given location of the image data and a second branch further localizes the curve within the given location if such a curve is present. Training of the different branches of the neural network may be decoupled.

System and method of determining a curve

A system to recognize objects in an image includes an object detection network outputs a first hierarchical-calculated feature for a detected object. A face alignment regression network determines a regression loss for alignment parameters based on the first hierarchical-calculated feature. A detection box regression network determines a regression loss for detected boxes based on the first hierarchical-calculated feature. The object detection network further includes a weighted loss generator to generate a weighted loss for the first hierarchical-calculated feature, the regression loss for the alignment parameters and the regression loss of the detected boxes. A backpropagator backpropagates the generated weighted loss. A grouping network forms, based on the first hierarchical-calculated feature, the regression loss for the alignment parameters and the bounding box regression loss, at least one of a box grouping, an alignment parameter grouping, and a non-maximum suppression of the alignment parameters and the detected boxes.

System and method for a unified architecture multi-task deep learning machine for object recognition

Apparatuses (including user equipment (UE) and modem chips for UEs), systems, and methods for UE downlink Hybrid Automatic Repeat reQuest (HARQ) buffer memory management are described. In one method, the entire UE DL HARQ buffer memory space is pre-partitioned according to the number and capacities of the UE's active carrier components. In another method, the UE DL HARQ buffer is split between on-chip and off-chip memory so that each partition and sub-partition is allocated between the on-chip and off-chip memories in accordance with an optimum ratio.

Growing community of inventors

La Jolla, CA, United States of America

Arvind Yedla