An image providing apparatus configured to generate, by using a first artificial intelligence (AI) network, AI metadata including class information and at least one class map, in which the class information includes at least one class corresponding to a type of an object among a plurality of predefined objects included in a first image and the at least one class map indicates a region corresponding to each class in the first image, generate an encoded image by encoding the first image, and output the encoded image and the AI metadata through the output interface.

A computer vision system configured for detection and recognition of objects in video and still imagery in a live or historical setting uses a teacher-student object detector training approach to yield a merged student model capable of detecting all of the classes of objects any of the teacher models is trained to detect. Further, training is simplified by providing an iterative training process wherein a relatively small number of images is labeled manually as initial training data, after which an iterated model cooperates with a machine-assisted labeling process and an active learning process where detector model accuracy improves with each iteration, yielding improved computational efficiency. Further, synthetic data is generated by which an object of interest can be placed in a variety of setting sufficient to permit training of models. A user interface guides the operator in the construction of a custom model capable of detecting a new object.

A computer vision system configured for detection and recognition of objects in video and still imagery in a live or historical setting uses a teacher-student object detector training approach to yield a merged student model capable of detecting all of the classes of objects any of the teacher models is trained to detect. Further, training is simplified by providing an iterative training process wherein a relatively small number of images is labeled manually as initial training data, after which an iterated model cooperates with a machine-assisted labeling process and an active learning process where detector model accuracy improves with each iteration, yielding improved computational efficiency. Further, synthetic data is generated by which an object of interest can be placed in a variety of setting sufficient to permit training of models. A user interface guides the operator in the construction of a custom model capable of detecting a new object.

A method for training a face recognition model includes: acquiring a plurality of first training images being uncovered face images, and acquiring a plurality of covering object images; generating a plurality of second training images by separately fusing the plurality of covering object images with the uncovered face images; and training the face recognition model by inputting the plurality of first training images and the plurality of second training images into the face recognition model.

A geometric pattern matching method and a device for performing the method includes determining, by a geometric pattern matching device, information on reference geometric pattern contour points for a learning object on a learning image. The method can further include determining, by the geometric pattern matching device, information on detection object contour points for a detection object on a detection image, and performing, by the geometric pattern matching device, geometric pattern matching between the learning object and the detection object on the basis of the information on the reference geometric pattern contour points and the information on the detection object contour points.

A geometric pattern matching method and a device for performing the method includes determining, by a geometric pattern matching device, information on reference geometric pattern contour points for a learning object on a learning image. The method can further include determining, by the geometric pattern matching device, information on detection object contour points for a detection object on a detection image, and performing, by the geometric pattern matching device, geometric pattern matching between the learning object and the detection object on the basis of the information on the reference geometric pattern contour points and the information on the detection object contour points.

A method for training a model based on knowledge distillation includes: inputting feature vectors obtained based on trained sample images into a first coding layer and a second coding layer, in which the first coding layer belongs to a first model, and the second coding layer belongs to a second model; obtaining first feature vectors by aggregating output results of the first coding layer; determining second feature vectors based on outputs of the second coding layer; and updating the first feature vectors by performing a distillation on the first feature vectors and the second feature vectors.

A method for training a model based on knowledge distillation includes: inputting feature vectors obtained based on trained sample images into a first coding layer and a second coding layer, in which the first coding layer belongs to a first model, and the second coding layer belongs to a second model; obtaining first feature vectors by aggregating output results of the first coding layer; determining second feature vectors based on outputs of the second coding layer; and updating the first feature vectors by performing a distillation on the first feature vectors and the second feature vectors.

A migration system of a learning model for cell image analysis is a system that migrates a learning model from a first learning device to a second learning device, in which the second learning device includes an algorithm consistency determination unit that determines, based on second algorithm specification information and first algorithm specification information, whether or not consistency is established between a first algorithm and a second algorithm, and a learning model parameter setting unit that sets a first parameter to be used together with the second algorithm.

A migration system of a learning model for cell image analysis is a system that migrates a learning model from a first learning device to a second learning device, in which the second learning device includes an algorithm consistency determination unit that determines, based on second algorithm specification information and first algorithm specification information, whether or not consistency is established between a first algorithm and a second algorithm, and a learning model parameter setting unit that sets a first parameter to be used together with the second algorithm.