A license plate classification method, a license plate classification apparatus and a computer-readable storage medium are provided. The method includes: performing a license plate recognition process on a first license plate image to obtain a license plate recognition result; performing an encoding process on the license plate recognition result to obtain a first license plate feature; performing a feature extraction process on the first license plate image to obtain a second license plate feature; and processing the first license plate feature and the second license plate feature through a classification network to obtain a first license plate classification result. In this way, an accuracy of license plate classification is improved.

Disclosed is a fusion network-based method for image super-resolution and non-uniform motion deblurring. The method achieves, for the first time, restoration of a low-resolution non-uniform motion-blurred image based on a deep neural network. The network uses two branch modules to respectively extract features for image super-resolution and non-uniform motion deblurring, and achieves, by means of a feature fusion module that is trainable, adaptive fusion of outputs of the two branch modules for extracting features. Finally, an upsampling reconstruction module achieves a non-uniform motion deblurring and super-resolution task. According to the method, a self-generated set of training data is configured to perform offline training on a network, thereby achieving restoration of the low-resolution non-uniform motion-blurred image.

An audio visual haptic signal reconstruction method includes first utilizing a large-scale audio-visual database stored in a central cloud to learn knowledge, and transferring same to an edge node; then combining, by means of the edge node, a received audio-visual signal with knowledge in the central cloud, and fully mining semantic correlation and consistency between modals; and finally fusing the semantic features of the obtained audio and video signals and inputting the semantic features to a haptic generation network, thereby realizing the reconstruction of the haptic signal. The method effectively solves the problems that the number of audio and video signals of a multi-modal dataset is insufficient, and semantic tags cannot be added to all the audio-visual signals in a training dataset by means of manual annotation. Also, the semantic association between heterogeneous data of different modals are better mined, and the heterogeneity gap between modals are eliminated.

Described herein is a mechanism for visual recognition of items or visual search using Optical Character Recognition (OCR) of text in images. Recognized OCR blocks in an image comprise position information and recognized text. The embodiments utilize a location-aware feature vector created using the position and recognized information in each recognized block. The location-aware features of the feature vector utilize position information associated with the block to calculate a weight for the block. The recognized text is used to construct a tri-character gram frequency, inverse document frequency (TGF-IDP) metric using tri-character grams extracted from the recognized text. Features in location-aware feature vector for the block are computed by multiplying the weight and the corresponding TGF-IDF metric. The location-aware feature vector for the image is the sum of the location-aware feature vectors for the individual blocks.

Disclosed is a method and system for predicting a touch interaction position on a large display based on a binocular camera. The method includes: separately acquiring arm movement video frames of a user and facial and eye movement video frames of the user by a binocular camera; extracting a video clip of each tapping action from the arm movement video frames and the facial and eye movement video frames and obtaining a key frame by screening; marking the key frame of each tapping action with coordinates to indicate coordinates of a finger in a display screen; inputting the marked key frame to an efficient convolutional network for online video understanding (ECO)-Lite neural network for training to obtain a predictive network model; and inputting a video frame of a current operation to be predicted to the predictive network model and outputting a touch interaction position predicted for the current operation.

An exemplary embodiment provides an intruder detection method capable of accurately detecting an intruder and estimating an abnormal behavior of the intruder even when viewpoints of acquired images are different from each other. An intruder detection method is suitable for being performed by an intruder detection device for detecting an intruder based on images and includes: receiving input images acquired by multiple cameras; extracting feature maps associated with a plurality of viewpoints by applying the input images to a plurality of convolutional neural networks provided separately for the plurality of viewpoints of the images; and detecting the intruder based on the feature maps associated with the plurality of viewpoints.

A system may be used to determined object characteristics and/or generate bounding boxes for objects in a vehicle scene by enriching later-in-time feature maps using earlier-in-time feature maps. The system may generate a feature map from a received. Using an earlier-in-time feature map, the system may enrich semantic data of the generated feature map to form an enriched feature map. The system may use the enriched feature map to generate one or more object characteristics of an object in the scene.

Methods, systems, and articles of manufacture to improve image recognition searching are disclosed. In some embodiments, a first document image of a known object is used to generate one or more other document images of the same object by applying one or more techniques for synthetically generating images. The synthetically generated images correspond to different variations in conditions under which a potential query image might be captured. Extracted features from an initial image of a known object and features extracted from the one or more synthetically generated images are stored, along with their locations, as part of a common model of the known object. In other embodiments, image recognition search effectiveness is improved by transforming the location of features of multiple images of a same known object into a common coordinate system. This can enhance the accuracy of certain aspects of existing image search/recognition techniques including, for example, geometric verification.

An identity recognition method includes: acquiring biometric pattern information describing a biometric pattern of a first object; performing feature extraction on the biometric pattern information to obtain a global pattern feature and a local pattern feature; fusing the global pattern feature and the local pattern feature to obtain a fused pattern feature of the first object; and performing identity recognition on the first object based on the fused pattern feature of the first object.

Provided are a lane line recognition method, an electronic device and a storage medium, relating to a technical field of artificial intelligence, in particular to technical fields of intelligent transportation, automatic driving and deep learning. The lane line recognition method includes: extracting a basic feature of an original image; recognizing at least one lane line node in the original image by using the basic feature of the original image; extracting a local feature from the basic feature of the original image by using the at least one lane line node; fusing the basic feature and the local feature; and recognizing a lane line in the original image based on a fused result.