Generating a machine learning model for objects based on augmenting the objects with physical properties

F Lecue, VO ANTONINO, S Hamiti, G Kaila - US Patent 10,565,475, 2020 - Google Patents
(57) ABSTRACT A device receives images of a video stream, models for objects in the
images, and physical property data for the objects, and maps the models and the physical …

Target detection method, apparatus, and system

Y Yang, Y Jiang, M Chen, S Yang - US Patent 11,367,272, 2022 - Google Patents
A target detection method and apparatus, in which the method includes: obtaining a target
candidate region in a to-be-detected image; determining at least two part candidate regions …

Image identification using neural networks

X Yang, Z Zheng, Z Yu - US Patent App. 16/357,047, 2020 - Google Patents
A neural network is trained to perform a re-identification task in which it is determined
whether one or more features present in a first image appear also in a second image. During …

Learning method and learning device for fluctuation-robust object detector based on CNN using target object estimating network adaptable to customers' requirements …

K Kim, Y Kim, I Kim, HK Kim, W Nam, S Boo… - US Patent …, 2019 - Google Patents
A method for learning parameters of an object detector by using a target object estimating
network adaptable to customers' requirements such as KPI is provided. When a focal length …

Method for acquiring sample images for inspecting label among auto-labeled images to be used for learning of neural network and sample image acquiring device …

K Kim, Y Kim, I Kim, HK Kim, W Nam, S Boo… - US Patent …, 2019 - Google Patents
A method for acquiring a sample image for label-inspecting among auto-labeled images for
learning a deep learning network, optimizing sampling processes for manual labeling, and …

Method and device for merging object detection information detected by each of object detectors corresponding to each camera nearby for the purpose of …

K Kim, Y Kim, I Kim, HK Kim, W Nam, S Boo… - US Patent …, 2019 - Google Patents
A method for merging object detection information detected by object detectors, each of
which corresponds to each of cameras located nearby, by using V2X-based auto labeling …

Road curvature generation in real-world images as a method of data augmentation

G Xinwei - US Patent 11,055,543, 2021 - Google Patents
US11055543B2 - Road curvature generation in real-world images as a method of data
augmentation - Google Patents US11055543B2 - Road curvature generation in real-world …

Learning method and learning device for adjusting parameters of CNN in which residual networks are provided for meta learning, and testing method and testing …

K Kim, Y Kim, I Kim, HK Kim, W Nam, S Boo… - US Patent …, 2019 - Google Patents
A CNN-based method for meta learning, ie, learning to learning, by using a learning device
including convolutional layers capable of applying convolution operations to an image or its …

Learning method, learning device using regression loss and testing method, testing device using the same

K Kim, Y Kim, I Kim, HK Kim, W Nam, S Boo… - US Patent …, 2019 - Google Patents
A method for learning parameters of a CNN based on regression losses is provided. The
method includes steps of: a learning device instructing a first to an n-th convolutional layers …

Post-processing method and device for detecting lanes to plan the drive path of autonomous vehicle by using segmentation score map and clustering map

K Kim, Y Kim, I Kim, HK Kim, W Nam, S Boo… - US Patent …, 2019 - Google Patents
A post-processing method for detecting lanes to plan the drive path of an autonomous
vehicle by using a segmentation score map and a clustering map is provided. The method …