Abstract: A method may include generating a first scaled variant pyramid associated with a first image and a second scaled variant pyramid associated with a second image; determining a pair of best-matching variants; detecting a first plurality of keypoints in the first variant and a second plurality of keypoints in the second variant; generating a descriptor for each keypoint of the first plurality of keypoints to create a first plurality of keypoint descriptors; generating a descriptor for each keypoint of the second plurality of keypoints to create a second plurality of keypoint descriptors; for each keypoint of the first plurality of keypoints, identifying a keypoint match from the second plurality of keypoints; determining that a number of keypoint matches is above a threshold number; and in response to determining that the number of keypoint matches is above the threshold number, determining that the first image matches the second image.

Abstract: A method and an apparatus for generating vehicle damage information are provided. The method includes: acquiring a to-be-processed vehicle image; for a target detection model in at least one pre-trained target detection model: inputting the vehicle image to the target detection model to generate a suspected damage area detection result; and determining a location of a suspected damage area in the vehicle image based on the generated suspected damage area detection result. A mechanism for detecting a suspected damage area is provided based on the target detection model, improving the vehicle damage assessment efficiency.

Abstract: In a method of using a polarimeter for improved mapping and perception of objects on the ground, the polarimeter records raw image data to obtain polarized images of an area. The raw image data is processed to form processed images. The processed images are enhanced, and objects are detected and tracked. The polarimeter may be in a vehicle on the ground or in the air.

Abstract: Implementations include receiving sample data, the sample data being generated as digital data representative of a sample of the product, providing a set of features by processing the sample data through multiple layers of a residual network, a first layer of the residual network identifying one or more features of the sample data, and a second layer of the residual network receiving the one or more features of the first layer, and identifying one or more additional features, processing the set of features using a CNN to identify a set of regions, and at least one object in a region of the set of regions, and determine a type of the at least one object, and selectively issuing an alert at least partially based on the type of the at least one object, the alert indicating contamination within the sample of the product.

Abstract: Systems and techniques are provided for recalibrating cameras in a real space for tracking puts and takes of items by subjects. The method includes first processing one or more selected images selected from a plurality of sequences of images received from a plurality of cameras calibrated using a set of calibration images that were used to calibrate the cameras previously. The first processing includes a process step to extract a plurality of feature descriptors from the images. The first processing also includes a process step to match one or more feature descriptors as extracted from the selected images with feature descriptors extracted from the set of calibration images that were used to calibrate the cameras previously. The feature descriptors correspond to points located at displays or structures that remain substantially immobile.

Abstract: An apparatus comprising a memory and one or more processing circuits is provided. The memory stores a blend table having blend weights. The processing circuits, for partitions of the blend table: determine whether a subset of the pixels associated with the partition includes pixels associated with seamlines defined in a three-dimensional surface representation of the scene. If none of the subset of the pixels are associated with the seamlines, the processing circuits populate a region of the virtual image corresponding to the partition with pixel values from an image captured by one of the plurality of image capture devices. If one or more of the subsets of the pixels is associated with the seamlines, the processing circuits populate the region of the virtual image associated with the partition with blended pixel values from two or more images captured by two or more of the plurality of image capture devices.

Abstract: Methods and apparatus to match images using semantic features are disclosed. An example apparatus includes a semantic labeler to determine a semantic label for each of a first set of points of a first image and each of a second set of points of a second image; a binary robust independent element features (BRIEF) determiner to determine semantic BRIEF descriptors for a first subset of the first set of points and a second subset of the second set of points based on the semantic labels; and a point matcher to match first points of the first subset of points to second points of the second subset of points based on the semantic BRIEF descriptors.

Abstract: A probabilistic model is provided based on an output of a matching procedure that matches a particular object to representations of objects, where the probabilistic model relates a probability of an object being present to a number of matching features. The probabilistic model is used for detecting whether a particular object is present in received visual data.

Abstract: This invention relates to means and processes of analysing cultural artefacts, for example to authenticate works of art or to reconstruct fragmented archaeological artefacts digitally, by first scanning the target artefact to produce a digital data model of the target artefact, which is then transformed algorithmically to obtain a target digital transform. This is analysed to identify and extract digital data minutiae from the digital transform data. Then, a number of comparator artefacts are scanned using the same scanning technologies to produce a digital data model of each comparator artefact. The same processes of algorithmic transformation and digital data minutiae extraction are applied to the comparator digital data models.

Abstract: An image processing system includes a processor and optical flow (OF) determination logic for quantifying relative motion of a feature present in a first frame of video and a second frame of video that provide at least one of temporally and spatially ordered images with respect to the two frames of video. The OF determination logic configures the processor to implement performing OF estimation between the first frame and second frame using a pyramidal block matching (PBM) method to generate an initial optical flow (OF) estimate at a base pyramid level having integer pixel resolution, and refining the initial OF estimate using at least one pass of a modified Lucas-Kanade (LK) method to provide a revised OF estimate having fractional pixel resolution.

Abstract: A method, a system and an electronic device for processing audio-visual data. In the method, a first dataset is obtained, where the first dataset includes several data pairs, and each of the data pairs in the first dataset includes a video frame and an audio clip that match each other. A multi-channel feature extraction network model is established to extract the visual features of each video frame and the auditory features of each audio clip in the first dataset. A contrastive loss function model is established using the extracted visual features and the auditory features to train the multi-channel feature extraction network. A classifier is established to determine whether an input audio-visual data pair is matched.

Abstract: An image labeling apparatus, method, and computer program product are provided. The image labeling apparatus generates several image features of several to-be-processed images and classifies them into several groups. For each group, the image labeling apparatus: (a) selects the image feature closest to the center as a representative feature and a compared feature, (b) selects the image feature farthest to the compared feature as a candidate feature, and (c) compares the similarity between the candidate feature and each representative feature with a threshold. When all the similarities are smaller than the threshold, adopts the candidate feature as another representative feature and the compared feature and repeats (b) and (c). The group(s) whose number of representative features is more than another threshold will be re-classified.

Abstract: A computer-implemented system and method for analyzing a biological sample are disclosed. A sequence of images of the biological sample from an image capture device are received from an image capture device, wherein each image of the sequence of images is acquired at a particular focal plane of the biological sample. In addition, an object map is developed from the sequence of images and the object map is analyzed to measure a three-dimensional characteristic of a particular object. The object map comprises a plurality of object map planes, each object map plane is associated with one image of the sequence of images, and pixels of the object map planes associated with the particular object in the sequence of images are assigned a unique identifier associated with the particular object.

Abstract: Described is a multiple-camera system and process for detecting, tracking, and re-verifying agents within a materials handling facility. In one implementation, a plurality of feature vectors may be generated for an agent and maintained as an agent model representative of the agent. When the object being tracked as the agent is to be re-verified, feature vectors representative of the object are generated and stored as a probe agent model. Feature vectors of the probe agent model are compared with corresponding feature vectors of candidate agent models for agents located in the materials handling facility. Based on the similarity scores, the agent may be re-verified, it may be determined that identifiers used for objects tracked as representative of the agents have been flipped, and/or to determine that tracking of the object representing the agent has been dropped.

Abstract: An imaging apparatus an image sensor which generates a reference image data of a reference area that includes a measuring area in which a display panel is positioned and an image processor which determines the measuring area by analyzing luminance of the reference image data and generates a crop image data that includes a measuring image data of the measuring area.

Abstract: The technology disclosed relates to authenticating users using a plurality of non-deterministic registration biometric inputs. During registration, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate sets of feature vectors. The non-deterministic biometric inputs can include a plurality of face images and a plurality of voice samples of a user. A characteristic identity vector for the user can be determined by averaging feature vectors. During authentication, a plurality of non-deterministic biometric inputs are given as input to a trained machine learning model to generate a set of authentication feature vectors. The sets of feature vectors are projected onto a surface of a hyper-sphere. The system can authenticate the user when a cosine distance between the authentication feature vector and a characteristic identity vector for the user is less than a pre-determined threshold.

Abstract: The present disclosure can provide a person re-identification apparatus and method that includes an identity feature extraction part configured to receive a multiple number of images each including a person requiring re-identification, extract features related to an identity of a person included in each image according to a pattern estimation method learned beforehand, and obtain an identity-related feature vector for each image; and a re-identification determination part configured to analyze a degree of similarity between an identity-related feature vector obtained for a base image including a search target from among the plurality of images and an identity-related feature vector obtained for another image to determine whether or not a person corresponding to the search target is included in the other image.

Abstract: Managing supply chain inventory is accomplished by measuring the dimensions and the weight of a product. A plurality of images of the product is obtained. The shape of the outer surface of the product is determined. The center of gravity of the product can be calculated. A product model that includes a three dimensional representation of the shape of a configuration of the product and optionally the center of gravity of the product is generated. The product model is stored in memory for future use.

Abstract: Disclosed are systems, devices, and methods for detecting characteristics of an unhatched egg. A set of images of an unhatched egg are obtained, where each of the spectral images is obtained in a particular wavelength range. The set of images is processed to extract image features, where the image features includes an image texture feature. The extracted image features are processed to classify the unhatched egg according to at least one characteristic. The at least one characteristic may include fertility and/or gender.

Abstract: The present invention discloses a visual relationship detection method based on a region-aware learning mechanism, comprising: acquiring a triplet graph structure and combining features after its aggregation with neighboring nodes, using the features as nodes in a second graph structure, and connecting in accordance with equiprobable edges to form the second graph structure; combining node features of the second graph structure with features of corresponding entity object nodes in the triplet, using the combined features as a visual attention mechanism and merging internal region visual features extracted by two entity objects, and using the merged region visual features as visual features to be used in the next message propagation by corresponding entity object nodes in the triplet; and after a certain number of times of message propagations, combining the output triplet node features and the node features of the second graph structure to infer predicates between object sets.