Methods, systems, articles of manufacture, and apparatus to recalibrate confidences for image classification are disclosed. An example apparatus to classify an image includes an image crop detector to detect a first image crop from the image, the first image crop corresponding to a first object, a grouping controller to select a second image crop corresponding to a second object at a location of the first object, a prediction generator to, in response to executing a trained model, determine a label corresponding to the first object and a confidence level associated with the label, and a confidence recalibrator to recalibrate the confidence level based on a probability of the first object having a first attribute based on the second object having a second attribute, the confidence level recalibrated to increase an accuracy of the image classification.

An item recognition system uses a top camera and one or more peripheral cameras to identify items. The item recognition system may use image embeddings generated based on images captured by the cameras to generate a concatenated embedding that describes an item depicted in the image. The item recognition system may compare the concatenated embedding to reference embeddings to identify the item. Furthermore, the item recognition system may detect when items are overlapping in an image. For example, the item recognition system may apply an overlap detection model to a top image and a pixel-wise mask for the top image to detect whether an item is overlapping with another in the top image. The item recognition system notifies a user of the overlap if detected.

Devices, systems, methods, and media are described for point cloud data augmentation using model injection, for the purpose of training machine learning models to perform point cloud segmentation and object detection. A library of surface models is generated from point cloud object instances in LIDAR-generated point cloud frames. The surface models can be used to inject new object instances into target point cloud frames at an arbitrary location within the target frame to generate new, augmented point cloud data. The augmented point cloud data may then be used as training data to improve the accuracy of a machine learned model trained using a machine learning algorithm to perform a segmentation and/or object detection task.

A system and method for confidence aided upsampling of categorical maps. In some embodiments, the method includes: determining a category of a first pixel of an image, the first pixel having a plurality of neighboring pixels, each of the neighboring pixels having a category; and processing the image based on the determined category. The determining may include: calculating a confidence weighted metric for each of the neighboring pixels, the confidence weighted metric being based on a maximum confidence value among each of the neighboring pixels; and determining the category of the first pixel based on the confidence weighted metric of each of the neighboring pixels and based on the category of one of the neighboring pixels.

Systems and methods of the present disclosure can include a computer-implemented method. The method can include obtaining a machine-learned model comprising one or more layers. At least a first layer of the one or more layers can be configured to receive a set of query vectors respectively associated with layer inputs, determine similarity measures the key vectors and the query vectors, apply a normalization operation to the plurality of respective similarity measures, and determine an output based on the normalized respective similarity measures and a plurality of class labels respectively associated with the plurality of key vectors.

Disclosed are a depth map processing method, a depth map processing apparatus, an electronic device, and a readable storage medium. The method includes: acquiring a depth map to be processed; acquiring a confidence level gradient and gradient angle of each pixel in the depth map; determining a candidate flying pixel and a target pixel according to the confidence level gradients, and generating a flying pixel depth threshold according to a depth value of the target pixel; acquiring depth values of a pair of pixels adjacent to the candidate flying pixel in a reference direction; and acquiring depth difference values, and determining, according to the depth difference values and the flying pixel depth threshold, whether the candidate flying pixel is a real flying pixel.

A method, system, and article of highly efficient neural network video image processing uses temporal correlations.

An integrated circuit (IC) is provided for in-situ anomaly detection. Sensors in the IC generates sensor datasets including information indicating conditions in the IC. A processing unit in the IC uses a sensor dataset and a model to detect and classify the anomaly. The processing unit may filter the sensor dataset, extract features from the filtered sensor dataset, and input the features into the model. The model outputs one or more classifications of the anomaly. A feature may be a distance vector that represents a difference between a data value in the filtered sensor dataset from a reference data value. The model may be a network of bit-cells in the IC. The model may be continuously trained in-situ, i.e., on the IC. The processing unit may provide the classifications to another processing unit in the IC. The other processing unit may mitigate the anomaly based on the classifications.

Feature descriptor matching is reformulated into a graph-matching problem. Keypoints from a query image and a reference image are initially matched and filtered based on the match. For a given keypoint, a feature graph is constructed based on neighboring keypoints surrounding the given keypoint. The feature graph is compared to a corresponding feature graph of a reference image for the matched keypoint. Relocalization data is obtained based on the comparison.

An information processing apparatus (100) includes a collation unit (102) that collates first feature information extracted from a person included in a first image (10) with first feature information indicating a feature of a retrieval target person, an extraction unit (104) that extracts second feature information from the person included in the first image in a case where a collation result in the collation unit (102) indicates a match, and a registration unit (106) that stores, in a second feature information storage unit (110), the second feature information extracted from the person included in the first image.