Abstract: A camera system comprises an image capturing device, object detection module, object tracking module, and match classifier. The object detection module receives image data and detects objects appearing in one or more of the images. The object tracking module temporally associates instances of detected objects, each of which has a signature representing features of the detected object. The match classifier matches object instances by analyzing data derived from the signatures. The match classifier determines whether the signatures match.

Abstract: Techniques for automatically detecting objects left behind. An example method includes receiving video frames of a scene from a three-dimensional (3D) camera and establishing, based on 3D depths associated with the video frames, a background of the scene. The method also includes detecting, in the frames, a foreground blob in the scene based on the foreground blob having a 3D depth that is different from the background. The method further includes detecting that the foreground blob has separated into a blob corresponding to a person and a second blob corresponding to an object, based on the person having a 3D depth that is different from the object. The method additionally includes determining that the person has been separated from the object for a threshold, and responsive to determining that the person has been separated from the object for the threshold, generating an alert indicating that the object is left behind.

Abstract: A method, system and computer program product for emulating depth data of a three-dimensional camera device is disclosed. The method includes concurrently operating the radar device and the 3D camera device to generate training radar data and training depth data respectively. Each of the radar device and the 3D camera device has a respective field of view. The field of view of the radar device overlaps the field of view of the 3D camera device. The method also includes inputting the training radar and depth data to the neural network. The method also includes employing the training radar and depth data to train the neural network. Once trained, the neural network is configured to receive real radar data as input and to output substitute depth data.

Abstract: Techniques for context aware access control with weapons detection are provided. An indication of an identity of a person is received at an access control system. The indication of the identity of the person includes a confidence level of the identification. An indication of a threat level of the person is received at a threat detection system. The threat level including a confidence level of the threat level. At least one of an identification threshold or a threat level threshold is modified based on the threat level confidence level or the confidence level of the identification. At least one of allowing access, allowing access with an alarm indication, or denying access to the person is based in part on the modified identification threshold or threat level threshold.

Abstract: One example temperature sensing device includes an electronic processor configured to receive a thermal image of a person captured by a thermal camera. The electronic processor is configured to determine a first temperature and a first location of a first hotspot on the person. The electronic processor is configured to determine a second location of a second hotspot on the person based on the second location being approximately symmetrical with respect to the first location about an axis, and the second hotspot having a second temperature that is approximately equal to the first temperature. The electronic processor is configured to determine a distance between the first location of the first hotspot and the second location of the second hotspot. In response to determining that the distance is within the predetermined range of distances, the electronic processor is configured to generate and output an estimated temperature of the person.

Abstract: Methods, systems, and techniques for tagless tracking of an object-of-interest are disclosed. Image and non-image data are generated across a plurality of camera-specific regions, and the object-of-interest is tracked over a global map formed as a composite of these regions.

Abstract: Methods, systems, and techniques for enhancing use of two-dimensional (2D) video analytics by using depth data. Two-dimensional image data representing an image comprising a first object is obtained, as well as depth data of a portion of the image that includes the first object. The depth data indicates a depth of the first object. An initial 2D classification of the portion of the image is generated using the 2D image data without using the depth data. The initial 2D classification is stored as an approved 2D classification when the initial 2D classification is determined consistent with the depth data. Additionally or alternatively, a confidence level of the initial 2D classification may be adjusted depending on whether the initial 2D classification is determined to be consistent with the depth data, or the depth data may be used with the 2D image data for classification.

Abstract: Methods, systems, and techniques for monitoring an object-of-interest within a region involve receiving at least data from two sources monitoring a region and correlating that data to determine that an object-of-interest depicted or represented in data from one of the sources is the same object-of-interest depicted or represented in data from the other source. Metadata identifying that the object-of-interest from the two sources is the same object-of-interest is then stored for later use in, for example, object tracking.

Abstract: One example temperature sensing device includes an electronic processor configured to receive a thermal image of a person captured by a thermal camera. The electronic processor is configured to determine a first temperature and a first location of a first hotspot on the person. The electronic processor is configured to determine a second location of a second hotspot on the person based on the second location being approximately symmetrical with respect to the first location about an axis, and the second hotspot having a second temperature that is approximately equal to the first temperature. The electronic processor is configured to determine a distance between the first location of the first hotspot and the second location of the second hotspot. In response to determining that the distance is within the predetermined range of distances, the electronic processor is configured to generate and output an estimated temperature of the person.

Abstract: A camera system comprises an image capturing device, object detection module, object tracking module, and match classifier. The object detection module receives image data and detects objects appearing in one or more of the images. The object tracking module temporally associates instances of detected objects, each of which has a signature representing features of the detected object. The match classifier matches object instances by analyzing data derived from the signatures. The match classifier determines whether the signatures match.

Abstract: A camera system comprises an image capturing device, object detection module, object tracking module, and match classifier. The object detection module receives image data and detects objects appearing in one or more of the images. The object tracking module temporally associates instances of detected objects, each of which has a signature representing features of the detected object. The match classifier matches object instances by analyzing data derived from the signatures. The match classifier determines whether the signatures match.

Abstract: There is provided an appearance search system comprising one or more cameras configured to capture video of a scene, the video having images of objects. The system comprises one or more processors and memory comprising computer program code stored on the memory and configured when executed by the one or more processors to cause the one or more processors to perform a method. The method comprises identifying one or more of the objects within the images of the objects. The method further comprises implementing a learning machine configured to generate signatures of the identified objects and generate a signature of an object of interest. The system further comprises a network configured to send the images of the objects from the camera to the one or more processors.

Abstract: Methods, systems, and techniques for classifying and/or detecting objects using visible and invisible light images. A visible light image and an invisible light image are received at a convolutional neural network (CNN). The visible light image depicts a region-of-interest imaged using visible light. The invisible light image depicts at least a portion of the region-of-interest imaged using invisible light, and at least one of the images depicts an object-of-interest within the portion of the region-of-interest shared between the images. The CNN then classifies and/or detects the object-of-interest using the images. The CNN may be trained to perform this classification and/or detection using pairs of visible and invisible light training images.

Abstract: Methods, systems, and techniques for generating two-dimensional (2D) and three-dimensional (3D) images and image streams. The images and image streams may be generated using active stereo cameras projecting at least one illumination pattern, or by using a structured light camera and a pair of different illumination patterns of which at least one is a structured light illumination pattern. When using an active stereo camera, a 3D image may be generated by performing a stereoscopic combination of a first set of images (depicting a first illumination pattern) and a 2D image may be generated using a second set of images (optionally depicting a second illumination pattern). When using a structured light camera, a 3D image may be generated based on a first image that depicts a structured light illumination pattern, and a 2D image may be generated from the first image and a second image that depicts a different illumination pattern.

Abstract: A method, system and computer program product for emulating depth data of a three-dimensional camera device is disclosed. The method includes concurrently operating the radar device and the 3D camera device to generate training radar data and training depth data respectively. Each of the radar device and the 3D camera device has a respective field of view. The field of view of the radar device overlaps the field of view of the 3D camera device. The method also includes inputting the training radar and depth data to the neural network. The method also includes employing the training radar and depth data to train the neural network. Once trained, the neural network is configured to receive real radar data as input and to output substitute depth data.

Abstract: Methods, systems, and techniques for generating two-dimensional (2D) and three-dimensional (3D) images and image streams. The images and image streams may be generated using active stereo cameras projecting at least one illumination pattern, or by using a structured light camera and a pair of different illumination patterns of which at least one is a structured light illumination pattern. When using an active stereo camera, a 3D image may be generated by performing a stereoscopic combination of a first set of images (depicting a first illumination pattern) and a 2D image may be generated using a second set of images (optionally depicting a second illumination pattern). When using a structured light camera, a 3D image may be generated based on a first image that depicts a structured light illumination pattern, and a 2D image may be generated from the first image and a second image that depicts a different illumination pattern.

Abstract: Methods, systems, and techniques for facilitating improved training of a supervised machine learning process, such as a decision tree. First and second object detections of an object depicted in a video are respectively generated using first and second object detectors, with the second object detector requiring more computational resources than the first object detector to detect the object. Whether a similarity and a difference between the first and second object detections respectively satisfy a similarity threshold and a difference threshold is determined. When the similarity threshold is satisfied, the first object detection is stored as a positive example for the machine learning training. When the difference threshold is satisfied, the first object detection is stored as a negative example for the machine learning training.

Abstract: Methods, systems, and techniques for facilitating improved training of a supervised machine learning process, such as a decision tree. First and second object detections of an object depicted in a video are respectively generated using first and second object detectors, with the second object detector requiring more computational resources than the first object detector to detect the object. Whether a similarity and a difference between the first and second object detections respectively satisfy a similarity threshold and a difference threshold is determined. When the similarity threshold is satisfied, the first object detection is stored as a positive example for the machine learning training. When the difference threshold is satisfied, the first object detection is stored as a negative example for the machine learning training.

Abstract: Methods, systems, and techniques for enhancing use of two-dimensional (2D) video analytics by using depth data. Two-dimensional image data representing an image comprising a first object is obtained, as well as depth data of a portion of the image that includes the first object. The depth data indicates a depth of the first object. An initial 2D classification of the portion of the image is generated using the 2D image data without using the depth data. The initial 2D classification is stored as an approved 2D classification when the initial 2D classification is determined consistent with the depth data. Additionally or alternatively, a confidence level of the initial 2D classification may be adjusted depending on whether the initial 2D classification is determined to be consistent with the depth data, or the depth data may be used with the 2D image data for classification.

Abstract: There are described methods and systems for facilitating identification of an object-of-interest. A face similarity score and a body similarity score of a query image relative to a gallery image are determined. A fused similarity score of the query image relative to the gallery image is determined by applying a relationship between the face similarity score, the body similarity score, and the fused similarity score. The fused similarity score is indicative of whether or not the object-of-interest and the potential object-of-interest are the same object-of-interest. For example, a machine learning process is used to fuse the face similarity score and the body similarity into the fused similarity score. The process is repeated for multiple gallery images. The gallery images may then be ranked according to their respective fused similarity scores.