A system may be configured to dynamically update deployed machine learning models. In some aspects, the system may receive sampled video information, generate first object detection information based on a cloud model and the sampled video information, and generate second object detection information based on a first edge model and the sampled video information. Further, the system may select, based on the first object detection information and the second object detection information, a plurality of training images from the sampled video information, detect motion information corresponding to motion of one or more detected objects within the plurality of training images, generate a plurality of annotated images based at least in part on the first object detection information and the motion information, and generate a second edge model based upon training the first edge model using the plurality of annotated images.

Apparatuses, systems, and techniques to train one or more neural networks. In at least one embodiment, one or more neural networks are trained to perform segmentation tasks based at least in part on training data comprising bounding box annotations.

An estimation device includes at least one processor, in which the processor functions as a learned neural network that derives a result of estimation relating to a bone density of a bone part from a simple radiation image acquired by simply imaging a subject including the bone part or a DXA scanning image acquired by imaging the subject by a DXA method. The learned neural network is learned by using, as teacher data, a composite two-dimensional image representing the subject, which is derived by combining a three-dimensional CT image of the subject, and information relating to the bone density of the subject.

An apparatus for recognizing an object in an image includes a preprocessing module configured to receive an image including an object and to output a preprocessed image by performing image enhancement processing on the received image to improve a recognition rate of the object included in the received image; and an object recognition module configured to recognize the object included in the image by inputting the preprocessed image to an input layer of an artificial neural network for object recognition.

The present invention relates to a method for improving reliability of artificial intelligence-based object recognition using collective intelligence-based mutual verification, in which: in a server interworking with an artificial intelligence module, one or more object regions included in learning data including a structured image or an unstructured image of a recognition target object to be recognized through the artificial intelligence module are recognized; object region recognition data which sets the recognized object regions is extracted and is provided to one or more user terminals in a designated order; a procedure for receiving, from the user terminals, object region selection data which selects at least one effective object region corresponding to the recognition target object among one or more object regions included in the object region recognition data is performed for a predetermined period of time; and the object region selection data of respective users received from the user terminals is mutually compared and analyzed by the same object region selection data, a collective intelligence-based mutual-verification procedure is performed, wherein the verification procedure is for selecting the object region selection data in which users of a designated ratio or more, among users who select the effective object region for each of the object region selection data, select the same effective object region, learning data to be input to the artificial intelligence module and to be learned in order to improve reliability of the artificial intelligence module is determined, and the learning data is input to the artificial intelligence module and is learned.

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately and efficiently learning parameters of a distilled neural network from parameters of a source neural network utilizing multiple augmentation strategies. For example, the disclosed systems can generate lightly augmented digital images and heavily augmented digital images. The disclosed systems can further learn parameters for a source neural network from the lightly augmented digital images. Moreover, the disclosed systems can learn parameters for a distilled neural network from the parameters learned for the source neural network. For example, the disclosed systems can compare classifications of heavily augmented digital images generated by the source neural network and the distilled neural network to transfer learned parameters from the source neural network to the distilled neural network via a knowledge distillation loss function.

The information processing apparatus obtains a distribution characteristic representing an attribute of a color included in an acquired first image group and determines the number of times of learning based on the distribution characteristic. The information processing apparatus generates a data set consisting of a set of the first image group and a second image group corresponding to the first image group and by using the generated data set, generates a learning model by performing learning using a network based on the determined number of times of learning.

Methods, systems, and apparatus, including computer programs encoded on computer storage media, for updating a knowledge distillation training system. One of the methods includes: providing, to a teacher model in a knowledge distillation training system, first data representing an image to cause the teacher model to generate teacher output data that indicates whether the image depicts an object of interest; providing, to a student model in the knowledge distillation training system, second data representing the image to cause the student model to generate student output data that indicates whether the image depicts an object of interest; determining whether an accuracy of the teacher output data satisfies an accuracy threshold; and in response to determining that the accuracy of the teacher output data does not satisfy the accuracy threshold: determining to skip updating the student model; and updating the student model using the student output data and ground truth data.

A method for updating a perception function of a vehicle having an Automated Driving System (ADS) is disclosed. The ADS has a self-supervised machine-learning (ML) algorithm for reconstructing an ingested image and a ML algorithm for an in-vehicle perception module for detecting one or more objects or free-space areas depicted in an ingested image. At first, an image of a scene in a surrounding environment of the vehicle is obtained. The obtained image is processed to obtain an output image with one or more detected objects or free-space areas. Then, an evaluation dataset is formed accordingly. The evaluation dataset and the obtained image is processed to obtain a reconstruction error value for each evaluation image and an evaluation image with highest reconstruction error value is selected among plurality of evaluation images. Using the obtained image and the selected evaluation image, the ML algorithm for the in-vehicle perception module is updated.

An information processing apparatus (100) includes a generation unit (132), a conversion unit (133), an evaluation unit (134), an image analysis unit (135), and a determination unit (138). The generation unit (132) performs supervised learning using first image data and teacher data and generates an image conversion model. The conversion unit (133) generates converted data from the second image data using the image conversion model. The evaluation unit (134) evaluates the converted data. The image analysis unit (135) analyzes second image data corresponding to the converted data, evaluation of which by the evaluation unit (134) is lower than a predetermined standard. The determination unit (138) determines, based on an analysis result by the image analysis unit (135), a photographing environment of photographing performed to acquire teacher data.