using pattern recognition or machine learning

Full Title

Arrangements for image or video recognition or understanding > using pattern recognition or machine learning

Classification Context

Section:

PHYSICS

Class:

COMPUTING OR CALCULATING; COUNTING

Subclass:

IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING

Scope Notes

Glossary: AE auto-encoder network AlexNet CNN designed by Alex Krizhevsky et al. Backprop backpropagation, an algorithm for adjusting the weights of an artificial neural network BERT bidirectional encoder representations from transformers, a transformer based artificial neural network C4.5 an algorithm for learning decision trees CART classification and regression trees CNN convolutional neural network, an artificial neural network that includes convolutional layers CPD coherent point drift, an algorithm for matching point clouds DAG directed acyclic graph DBSCAN density-based spatial clustering of applications with noise, a non-parametric clustering algorithm which does not require specifying the number of clusters in advance. DNN deep neural network EMD earth mover’s distance/Wasserstein metric FCL fully connected layer of an artificial neural network FCNN fully convolutional neural network GAN generative adversarial network GMM Gaussian mixture model GoogLeNet deep convolutional neural network ICA independent component analysis ICP iterative closest point, an algorithm for matching point clouds ID3 iterative Dichotomiser 3, an algorithm for learning decision trees Inception convolutional neural network which concatenates several filters of different sizes at the same level of the network. IoU intersection over union, a measure for quantifying the accuracy of an object detection algorithm KDE kernel density estimation, an algorithm for estimating the probability density function of a random variable kernel function which expresses an inner product of two inputs in another feature space. KLT Karhunen-Loève transform K-Means data clustering algorithm KNN K-nearest neighbour; a classification algorithm which, for a given data sample, chooses the k most similar samples from a training set, retrieves their respective class labels, and assigns a class label to the data sample by majority decision; variant: 1NN, which is KNN for k=1. LASSO least absolute shrinkage and selection operator LDA linear discriminant analysis LeNet early CNN that firstly demonstrated the performance of CNNs on handwritten character recognition. LSTM long short-term memory, a recurrent neural network LVQ learning vector quantisation MDS multi-dimensional scaling MLP multi-layer perceptron MRF Markov random field MS COCO annotated image dataset overfitting trained model suffers from overfitting if it performs well on the training data, but generalises poorly on new test data. PASCAL VOC collection of datasets for object detection PCA principal component analysis PDF probability density function Perceptron simple feed-forward neural network RANSAC random sample consensus, a popular regression algorithm RBF radial basis function Res-Net residual neural network, an artificial neural network having shortcuts / skip connections between different layers R-CNN convolutional neural network using a region proposal algorithm for object detection (variants: fast R-CNN, faster R-CNN, cascade R-CNN) ROC receiver-operating characteristics RPM robust point matching, an algorithm for matching point clouds RVM relevance vector machine SOM self-organising maps, an algorithm for generating a low-dimensional representation of data while preserving the topological structure of the data SSD single shot (multibox) detector, a neural network for object detection SVD singular value decomposition SVM support vector machine test data data set different from the training data, used for testing the performance of a trained model training data data set used for adjusting the parameters of the model during training transformer transformers deep learning model that uses attention to give different weights to individual parts of the input data. U-Net neural network having a specific layer structure validation data data set used for testing the performance of the model during training YOLO you only look once, an artificial neural network used for object detection (comes in various versions: YOLO v2, YOLO v3 etc.) | Limiting references: Pattern recognition performed by an arrangement of optical devices rather than by machine learning | Application references: Scenes; Scene-specific elements Character recognition Image or video recognition or understanding of human-related, animal-related or biometric patterns in image or video data

Child Classifications

Navigate with arrow keys, Enter to open

• G06V 10/72 Data preparation, e.g. statistical preprocessing of image or video features
• G06V 10/74 Image or video pattern matching; Proximity measures in feature spaces
• G06V 10/762 using clustering, e.g. of similar faces in social networks
• G06V 10/764 using classification, e.g. of video objects
• G06V 10/766 using regression, e.g. by projecting features on hyperplanes
• G06V 10/77 Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
• G06V 10/82 using neural networks
• G06V 10/84 using probabilistic graphical models from image or video features, e.g. Markov models or Bayesian networks
• G06V 10/86 using syntactic or structural representations of the image or video pattern, e.g. symbolic string recognition; using graph matching