Samsung Electronics Co., Ltd.

Korea Advanced Institute of Science and Technology

Dokwan Oh

Dae Hyun Ji

Cheolhun Jang

Jaewoo Lee

Yoonsuk Hyun

DongWook Lee

Jae-young Choi

Seungjun Shin

JaeWook Yoo

Wonju Lee

Nayeon Kim

Woosup Han

Yong-man Ro

Jahoo Koo

Choong Hwan Choi

Suji Kim

Moonsub Byeon

Dongwon Jang

Dasol Han

Sujin Jang

Joohan Na

Dae-hoe Kim

A method with image transformation includes: identifying an original image; and determining a transformed image by inputting the original image to a neural network model configured to transform a color of the original image, wherein the neural network model comprises an operation block configured to perform white balancing on the original image, a correction block configured to correct a color of an output image of the operation block, and a mapping block configured to apply a lookup table to an output image of the correction block.

Method and apparatus with image transformation

Disclosed are a method and apparatus for training a neural network model to increase performance of the neural network model, the method including receiving input data and target data, pooling, by a neural network model, on a feature map extracted from the input data based on a probability for each of classes of the feature map, generating output data by inputting the input data to a neural network model, determining a loss based on comparing the output data and the target data and an auxiliary loss of the pooling, and training the neural network model based on the loss.

Method and apparatus for training neural network models to increase performance of the neural network models

A method of generating lane information using a neural network includes generating a lane probability map based on an input image, generating lane feature information and depth feature information by applying the lane probability map to a second neural network, generating depth distribution information by applying the depth feature information to a third neural network, generating spatial information based on the lane feature information and the depth distribution information, generating offset information including a displacement between a position of a lane and a reference line by applying the spatial information to a fourth neural network, and generating three-dimensional (3D) lane information using the offset information.

Method and apparatus with lane generation

An electronic device extracts feature data from an input image, calculates one or more class maps from the feature data using a classifier layer, calculates one or more cluster maps from the feature data using a clustering layer, and generates image segmentation data using the one or more class maps and the one or more cluster maps.

Image segmentation method and device

A processor-implemented method with data processing using a neural network includes: determining a first translated image by translating a first image based on a second image, the first image and a second image that having different distortions, such that a distortion of the first image corresponds to a distortion of the second image; determining a first retranslated image by translating the first translated image such that a distortion of the first translated image corresponds to a distortion of the first image; and training a first deformation field generator configured to determine a first relative deformation field that represents a relative deformation from the first image to the second image and a second deformation field generator configured to determine a second relative deformation field that represents a relative deformation from the second image to the first image, based on a loss between the first retranslated image and the first image.

Method and device with data processing using neural network

Disclosed is a training data generating apparatus and method to change output data inappropriate for training a student model to an ignore value such that the inappropriate output data is not used to train the student model, change output data appropriate for training the student model such that the student model outputs an improved result in comparison to output data of a teacher model, and change a label value in a form of probability to an identifier corresponding to intervals divided based on threshold label values using the teacher model based on input data.

Method and apparatus for generating training data to train student model using teacher model

An image obtaining apparatus, for obtaining an image from an image sensor provided in a vehicle, is provided. The image obtaining apparatus includes: a processor, and a memory including at least one instruction executable by the processor. In response to the at least one instruction being executed by the processor, the processor is configured to: obtain an initial image from the image sensor, the initial image including an object; determine a type of the object included in the initial image; adjust a shutter speed of the image sensor based on the determined type of the object included in the initial image; and obtain a first image from the image sensor based on the adjusted shutter speed.

Image obtaining method and apparatus

A method and apparatus for adjusting a neural network that classifies a scene of an input image into at least one class is provided. The method generates a feature image having a size that is less than a size of an input image by applying a convolutional network to the input image, determines at least one class corresponding to the feature image, generates a class image having a size corresponding to the size of the input image by applying a deconvolutional network to the feature image, calculates a loss of the class image based on a verification class image preset with respect to the input image, and adjusts the neural network based on the loss.

Image processing method and apparatus with neural network adjustment

A method of obtaining an image from an image sensor provided in a vehicle includes obtaining a vehicle speed of the vehicle; adjusting a shutter speed of the image sensor; adjusting either one or both of an international organization for standardization sensitivity (ISO) of the image sensor and an aperture of the image sensor based on the vehicle speed; and obtaining the image from the image sensor based on the adjusted shutter speed and the adjusted either one or both of the ISO and the aperture.

Disclosed is a method and apparatus that includes acquiring a driving image; acquiring positioning information indicating a location of a vehicle; extracting map information corresponding to the positioning information; determining a regression line function corresponding to a road on which the vehicle is traveling based on the map information; detecting the linearity of the road from the driving image using the regression line function; and indicating the detected linearity.

Method and apparatus with linearity detection

A method and apparatus for controlling a vision sensor are provided. The apparatus and corresponding method are configured to predict an expected point, on a traveling path of a host vehicle, at which an illumination variation greater than or equal to a threshold is expected to occur, and determine whether the host vehicle is located within a threshold distance. The apparatus and corresponding method are also configured to control a vision sensor in the host vehicle based on the expected illumination variation in response to the host vehicle being located within the threshold distance.

Method and apparatus for controlling vision sensor for autonomous vehicle

A training data generation method includes acquiring an image captured from a vehicle and location information of the vehicle corresponding to the image; acquiring map data corresponding to the acquired location information; determining truth data including information associated with a road included in the image from the acquired map data; and generating training data including the image and the determined truth data.

Apparatus and method for generating training data to train neural network determining information associated with road included in image

An apparatus for estimating a malignant tumor includes: a segmentor configured to segment a first medical image of an object into a first ROI including a mass and a second medical image of the object into a second ROI including a mass; an interest region determiner configured to acquire a first matched ROI and a second matched ROI, which is matched to the first matched ROI based on location information of the first ROI and the second ROI, respectively; a feature extractor configured to extract a similar feature indicating a degree of similarity between the first matched ROI and the second matched ROI, from the first matched ROI and the second matched ROI; and a classifier configured to generate malignant tumor information indicating whether the mass included in the first matched ROI and the second matched ROI is the malignant tumor, based on the extracted similar feature.

Growing community of inventors

Hwaseong-si, South Korea

Dokwan Oh