A method for face recognition by using a multiple patch combination based on a deep neural network is provided. The method includes steps of: a face-recognizing device, (a) if a face image with a 1-st size is acquired, inputting the face image into a feature extraction network, to allow the feature extraction network to generate a feature map by applying convolution operation to the face image with the 1-st size, and to generate multiple features by applying sliding-pooling operation to the feature map, wherein the feature extraction network has been learned to extract a feature using a face image for training having a 2-nd size and wherein the 2-nd size is smaller than the 1-st size; and (b) inputting the multiple features into a learned neural aggregation network, to allow the neural aggregation network to aggregate the multiple features and to output an optimal feature for the face recognition.

A method for face recognition by using a multiple patch combination based on a deep neural network is provided. The method includes steps of: a face-recognizing device, (a) if a face image with a 1-st size is acquired, inputting the face image into a feature extraction network, to allow the feature extraction network to generate a feature map by applying convolution operation to the face image with the 1-st size, and to generate multiple features by applying sliding-pooling operation to the feature map, wherein the feature extraction network has been learned to extract a feature using a face image for training having a 2-nd size and wherein the 2-nd size is smaller than the 1-st size; and (b) inputting the multiple features into a learned neural aggregation network, to allow the neural aggregation network to aggregate the multiple features and to output an optimal feature for the face recognition.

A learning method for supporting a safer autonomous driving through a fusion of information acquired from images and communications is provided. And the method includes steps of: (a) a learning device instructing a first neural network and a second neural network to generate an image-based feature map and a communication-based feature map by using a circumstance image and circumstance communication information; (b) the learning device instructing a third neural network to apply a third neural network operation to the image-based feature map and the communication-based feature map to generate an integrated feature map; (c) the learning device instructing a fourth neural network to apply a fourth neural network operation to the integrated feature map to generate estimated surrounding motion information; and (d) the learning device instructing a first loss layer to train parameters of the first to the fourth neural networks.

A key point detection method includes: obtaining first feature maps of a plurality of scales for an input image, scales of the first feature maps having a multiple relationship; performing forward processing on each first feature map through a first pyramid neural network to obtain second feature maps in one-to-one correspondence to the first feature maps, each second feature map having the same scale as that of its respective first feature map; performing reverse processing on each second feature map through a second pyramid neural network to obtain third feature maps in one-to-one correspondence to the second feature maps, each third feature map having the same scale as that of its respective second feature map; and performing feature fusion processing on each third feature map, and obtaining the position of each key point in the input image through the feature map subjected to the feature fusion processing.

Systems and methods for estimating a height of an object from a monocular image are described herein. Objects are detected in the image, each object being indicated by a region of interest. The image is then cropped for each region of interest and the cropped image scaled to a predetermined size. The cropped and scaled image is then input into a convolutional neural network (CNN), the output of which is an estimated height for the object. The height may be represented by a mean of a probability distribution of possible sizes, a standard deviation, as well as a level of confidence. A location of the object may be determined based on the estimated height and region of interest. A ground truth dataset may be generated for training the CNN by simultaneously capturing a LIDAR sequence with a monocular image sequence.

According to the embodiments of the present application, there are proposed a method and electronic device for retrieving an image, and computer readable storage medium. The method includes: processing an image to be retrieved using a first neural network to determine a local feature vector of the image to be retrieved; processing the image to be retrieved using a second neural network to determine a global feature vector of the image to be retrieved; and determining, based on the local feature vector and the global feature vector, an image having a similarity to the image to be retrieved which is higher than a similarity threshold.

An anomaly detection apparatus generates pieces of image data using a generator and train the generator and a discriminator that discriminates whether an image data, generated by the generator, is real or fake. The anomaly detection apparatus trains the generator such that the generator, in generating the pieces of image data to maximize the discrimination error of the discriminator, generate at least a piece of specified image data to reduce the discrimination error at a fixed rate with respect to the pieces of image data and trains, based on the pieces of image data and the at least a piece of specified image data, the discriminator to minimize the discrimination error.

An apparatus and method are disclosed for user and moving vehicle detection in which sensor data for a portable device is processed to determine whether the portable device is in a moving vehicle. Following a determination the portable device is in a moving vehicle, the sensor data is to characterize an association between the user and the portable device to determine whether the portable device is connected to the user. If the user is connected to the portable device, it is then determined if the portable device is being held in hand. If the portable device is held in hand, it is then determined if the user is operating the moving vehicle. Output from an image sensor of the portable device may be used in determining if the user is the operator.

A system for correlating sensor data in a vehicle includes a first sensor disposed on the vehicle to detect a plurality of first objects. A first object identification controller analyzes the first data stream, identifies the first objects, and determines first characteristics associated therewith. A second sensor disposed on the vehicle detects a plurality of second objects. A second object identification controller analyzes the second data stream, identifies the second objects, and determines second characteristics associated therewith. A model generator includes a plausibility voter to generate an environmental model of the objects existing in space around the vehicle. The model generator may use ASIL decomposition to provide a higher ASIL level than that of any of the sensors or object identification controllers alone. Matchings between uncertain objects are accommodated using matching distance probability functions and a distance-probability voter. A method of operation is also provided.

A method of optical and microwave synergistic retrieval of aboveground biomass, the method including: 1) obtaining an observation value of aboveground biomass (AGB) of a sample plot; 2) pre-processing laser radar (LiDAR) data, optical remote sensing data and microwave remote sensing data covering a research region, to yield crown height model (CHM) data, surface reflectance data and a backscattering coefficient, respectively; 3) extracting different LiDAR variables, extracting a plurality of optical characteristic vegetation indexes, and extracting a plurality of microwave characteristic variables; 4) establishing a multiple stepwise linear regression model of the biomass; 5) taking the biomass value of the LiDAR data coverage region as a training set and a verification sample set, and selecting samples for modeling and verification; 6) screening out the optical and microwave characteristic variables; and 7) constructing an optical model, a microwave model, and an optical and microwave synergistic model of AGB retrieval, respectively.