Qualcomm Incorporated

Hyun-Mook Cho

Taesu Kim

Duck Hoon Kim

Kisun You

Hyung-Il Koo

Minho Jin

Hee-Seok Lee

Te-Won Lee

Seungwoo Yoo

Duck-hoon Kim

Young-Ki Baik

Jeong-Kyun Lee

Kyuwoong Hwang

ChaeSeong Lim

Xinzhou Wu

Ananthapadmanabhan Arasanipalai Kandhadai

Kyu Woong Hwang

Pengjun Huang

Jayakrishnan Unnikrishnan

Ravi Teja Sukhavasi

Avdhut Joshi

Yongwoo Cho

Heesoo Myeong

Yoga Y Nadaraajan

Amir Salimi

Gautam Sachdeva

Shivam Agarwal

Sree Sesha Aravind Vadrevu

Jimin Jeon

Seong Ill Park

Disclosed are techniques for estimating a 3D bounding box (3DBB) from a 2D bounding box (2DBB). Conventional techniques to estimate 3DBB from 2DBB rely upon classifying target vehicles within the 2DBB. When the target vehicle is misclassified, the projected bounding box from the estimated 3DBB is inaccurate. To address such issues, it is proposed to estimate the 3DBB without relying upon classifying the target vehicle.

Bounding box estimation and object detection

Disclosed are techniques for determining a motion state of a target object. In an aspect, an on-board computer of an ego vehicle detects the target object in one or more images, determines one or more first attributes of the target object based on measurements of the one or more images, determines one or more second attributes of the target object based on measurements of a map of a roadway on which the target object is travelling, and determines the motion state of the target object based on the one or more first attributes and the one or more second attributes of the target object.

Determining a motion state of a target object

Certain aspects of the present disclosure are directed to methods and apparatus for deep learning in an artificial neural network. One example method generally includes receiving input data at an input to a layer of the neural network; replicating a group of neural processing units in the layer to form a superset of neural processing units, the superset comprising n instances of the group of neural processing units; processing the input data using the superset to generate output data for the layer; and determining an uncertainty of the output data. Processing the input data includes performing a dropout function by zeroing out one or more weights of a set of weights for each of the n instances of the superset of neural processing units and convolving, for each of the n instances in parallel, the input data with one or more non-zeroed out weights of the set of weights.

Efficient dropout inference for bayesian deep learning

A method performed by an apparatus is described. The method includes receiving remote data. The method also includes adapting model processing based on the remote data. The method further includes determining a driving decision based on the adapted model processing. In some examples, adapting model processing may include selecting a model, adjusting model scheduling, and/or adjusting a frame computation frequency.

Systems and methods for adaptive model processing

A method, which is performed by an electronic device, of automatically activating a flash for an image sensor of the electronic device is disclosed. The method may include receiving a first image including at least one text region and determining feature data characterizing the at least one text region in the first image. The method may also identify at least one candidate specular reflection region in the first image. Based on the feature data and the at least one candidate specular reflection region, the flash may be activated. Upon activating the flash, a second image including the at least one text region may be captured.

Activating flash for capturing images with text

A method, which is performed by an electronic device, for adjusting at least one image capturing parameter in a preview mode is disclosed. The method may include capturing a preview image of a scene including at least one text object based on a set of image capturing parameters. The method may also identify a plurality of text regions in the preview image. From the plurality of text regions, a target focus region may be selected. Based on the target focus region, the at least one image capturing parameter may be adjusted.

Image capturing parameter adjustment in preview mode

A method for responding in an augmented reality (AR) application of a mobile device to an external sound is disclosed. The mobile device detects a target. A virtual object is initiated in the AR application. Further, the external sound is received, by at least one sound sensor of the mobile device, from a sound source. Geometric information between the sound source and the target is determined, and at least one response for the virtual object to perform in the AR application is generated based on the geometric information.

Augmented reality with sound and geometric analysis

A method for recognizing a text block in an object is disclosed. The text block includes a set of characters. A plurality of images of the object are captured and received. The object in the received images is then identified by extracting a pattern in one of the object images and comparing the extracted pattern with predetermined patterns. Further, a boundary of the object in each of the object images is detected and verified based on predetermined size information of the identified object. Text blocks in the object images are identified based on predetermined location information of the identified object. Interim sets of characters in the identified text blocks are generated based on format information of the identified object. Based on the interim sets of characters, a set of characters in the text block in the object is determined.

System and method for recognizing text information in object

A method for processing a multi-channel image is disclosed. The method includes generating a plurality of grayscale images from the multi-channel image. At least one text region is identified in the plurality of grayscale images and text region information is determined from the at least one text region. The method generates text information of the multi-channel image based on the text region information. If the at least one text region includes a plurality of text regions, text region information from the plurality of text regions is merged to generate the text information. The plurality of the grayscale images is processed in parallel. In identifying the at least one text region, at least one candidate text region may be identified in the plurality of grayscale images and the at least one text region may be identified in the identified candidate text region.

Parallel processing method and apparatus for determining text information from an image

A method includes receiving an alarm sound including information related to an emergency event. The method also includes transmitting, to a server, identification information of the mobile device and the information. The method further includes receiving, from the server, an instruction for responding to the emergency event. The method further includes outputting the instruction.

Emergency handling system using informative alarm sound

A method for controlling an application in a mobile device is disclosed. The method includes receiving environmental information, inferring an environmental context from the environmental information, and controlling activation of the application based on a set of reference models associated with the inferred environmental context. In addition, the method may include receiving a sound input, extracting a sound feature from the sound input, transmitting the sound feature to a server configured to group a plurality of mobile devices into at least one similar context group, and receiving, from the server, information on a leader device or a non-leader device and the at least one similar context group.

Controlling applications in a mobile device based on environmental context

Embodiments of the invention describe methods and apparatus for performing context-sensitive OCR. A device obtains an image using a camera coupled to the device. The device identifies a portion of the image comprising a graphical object. The device infers a context associated with the image and selects a group of graphical objects based on the context associated with the image. Improved OCR results are generated using the group of graphical objects. Input from various sensors including microphone, GPS, and camera, along with user inputs including voice, touch, and user usage patterns may be used in inferring the user context and selecting dictionaries that are most relevant to the inferred contexts.

Camera OCR with context information

Methods, apparatuses, systems, and computer-readable media for rejecting false positive detection and tracking of image objects are presented. According to one or more aspects, a computing device may implement embodiments of the invention to use the movement of the mobile device for distinguishing false positives from true movement of the mobile device depicted in the field of view of the camera. In one embodiment, the actual movement of the mobile device may be measured using multi-modal sensor data from inertial sensors such as accelerometers and gyroscopes. In another embodiment, the actual movement of the device is calculated using the global movement of the mobile phone with reference to other objects in the field of view of the camera.

Growing community of inventors

Seoul, South Korea

Hyun-Mook Cho