A method for detecting at least one linear object in an input image is disclosed. The input image, and/or an extract of the input image, is fed to an image classifier, which classifies specified regions of the input image or extract in each case at least into relevant regions, the center of which lies in fairly close proximity to the center point of at least one linear object passing at least partially through this region, and background regions where this is not the case. For the relevant regions, coordinates are acquired from a regression which indicate at least one local course of the linear object in the relevant regions. From these coordinates, the course of the linear object is evaluated in the entire input image.

The present invention relates to dimensionality reduction for machine learning (ML) models. Herein are techniques that individually rank features and combine features based on their rank to achieve an optimal combination of features that may accelerate training and/or inferencing, prevent overfitting, and/or provide insights into somewhat mysterious datasets. In an embodiment, a computer ranks features of datasets of a training corpus. For each dataset and for each landmark percentage, a target ML model is configured to receive only a highest ranking landmark percentage of features, and a landmark accuracy achieved by training the ML model with the dataset is measured. Based on the landmark accuracies and meta-features values of the dataset, a respective training tuple is generated for each dataset. Based on all of the training tuples, a regressor is trained to predict an optimal amount of features for training the target ML model.

Among other things, the techniques described herein include a method for receiving a plurality of images of one or more containers while the one or more containers are being emptied, the plurality of images comprising a training set of images and a validation set of images; labeling each image of the plurality of images as including either an overfilled container or a not-overfilled container; processing each image of the plurality of images to reduce bias of a machine learning model; training, and based on the labeling, the machine learning model using the plurality of images; and optimizing the machine learning model by performing learning against the validation set, the optimized machine learning model being used to generate a prediction for a new image of a container, the prediction indicating whether the container in the new image was overfilled prior to the new container being emptied.

An object detection device (1) includes an object detection unit (2) that detects an object from an image including the object by neural computation using a CNN. The object detection unit (2) includes: a feature amount extraction unit (2a) that extracts a feature amount of the object from the image; an information acquisition unit (2b) that obtains a plurality of object rectangles indicating candidates for the position of the object on the basis of the feature amount and obtains information and a certainty factor of a category of the object for each of the object rectangles; and an object tag calculation unit (2c) that calculates, for each of the object rectangles, an object tag indicating which object in the image the object rectangle is linked to, on the basis of the feature amount. The object detection device (2) further includes an excess rectangle suppression unit (4) that separates a plurality of object rectangles for which a category of the object is the same into a plurality of groups according to the object tags, and deletes an excess object rectangle in each of the separated groups on the basis of the certainty factor.

Examples disclosed herein may involve (i) obtaining a first layer of map data associated with sensor data capturing a geographical area, the first layer of map data comprising an aggregated overhead-view image of the geographical area, where the aggregated overhead-view image is generated from aggregated pixel values from a plurality of images associated with the geographical area, (ii) obtaining a second layer of map data, the second layer of map data comprising label data for the geographical area derived from the aggregated overhead-view image of the geographical area, and (iii) causing the first layer of map data and the second layer of map data to be presented to a user for curation of the label data.

A method for generating an object detection dataset and a computer implemented object detection system are disclosed. The method comprises:

receiving a training image dataset comprising a plurality of images that include objects of interest, each image comprising pixel values corresponding to an imaged material generated by a penetrating imager;

generating a thresholded image for each of the plurality of images;

segmenting each thresholded image into images corresponding to objects;

creating a greyscale image per object from the segmented images corresponding to that object by, for each object, calculating an average pixel value for each pixel of the object from corresponding pixels of the object in the segmented images;

forming a greyscale image for the object from the averaged pixels;

storing the greyscale images in a data repository as an object detection dataset.

A learning processing device that is based on a neural network model and image data obtained by capturing an image of the object to be inspected, and constructs the neural network model used for inspecting the object to be inspected. The learning processing device is provided with a learning unit which performs a learning process under a prescribed learning condition on the basis of a list of the image data including a plurality of learning images and constructs the neutral network model. The learning unit embeds unique model identification data in the neural network model, whenever the neural network model is constructed.

Provided is a method for performing accurate object recognition in a stable manner in consideration of changes in a shooting environment. In such a method, a camera captures an image of a shooting location where an object is to be placed and an object included in an image of the shooting location is recognized utilizing a machine learning model for object recognition. The method further involves: determining necessity of an update operation on the machine learning model for object recognition at a predetermined time; when the update operation is necessary, causing the camera to capture an image of the shooting location where no object is placed to thereby re-acquire a background image for training; and causing the machine learning model to be trained using a composite image of a backgroundless object image and the re-acquired background image for training as training data.

A latent code defined in an input space is processed by the mapping neural network to produce an intermediate latent code defined in an intermediate latent space. The intermediate latent code may be used as appearance vector that is processed by the synthesis neural network to generate an image. The appearance vector is a compressed encoding of data, such as video frames including a person's face, audio, and other data. Captured images may be converted into appearance vectors at a local device and transmitted to a remote device using much less bandwidth compared with transmitting the captured images. A synthesis neural network at the remote device reconstructs the images for display.

A method for generating positive and negative training samples is presented. The method includes identifying false positive images of an object based on multiple images of an environment. The method also includes generating positive training samples from a set of images of the object. The method further includes generating a negative training sample from the false positive image. The method still further includes training an object detection system based on the positive training samples and the negative training sample.