Abstract: Various implementations disclosed are for detecting moving objects that are in a field of view of a head-mountable device (HMD). In various implementations, the HMD includes a display, an event camera, a non-transitory memory, and a processor coupled with the display, the event camera and the non-transitory memory. In some implementations, the method includes synthesizing a first optical flow characterizing one or more objects in a field of view of the event camera based on depth data associated with the one or more objects. In some implementations, the method includes determining a second optical flow characterizing the one or more objects in the field of view of the event camera based on event image data provided by the event camera. In some implementations, the method includes determining that a first object of the one or more objects is moving based on the first optical flow and the second optical flow.

Abstract: An enhanced flight vision system for an aircraft includes an image acquisition system configured to acquire images of the surroundings outside the aircraft and a display system configured to receive images produced by the image acquisition system and to display these images on a display in the cockpit of the aircraft. The display system is configured to acquire information about the flight path angle of the aircraft when approaching a runway, to calculate a difference between the flight path angle of the aircraft and a nominal angle and to deactivate the display of the images received from the image acquisition system on the display when the absolute value of the difference is greater than a first angular value.

Abstract: A rugged integrated helmet vision system is disclosed. The system includes an accessory connector for connecting an accessory such as a heads-up display (HUD) device to a helmet worn by a user. The accessory connector includes a connector arm assembly with a portion that attaches the connector to the helmet, and another portion that attaches the accessory to the connector. The portions each provide multiple degrees of freedom of movement to allow adjustment of the connector with respect to the helmet, and adjustment of the accessory with respect to an eye of the user. The connector also provides the ability to lock positions of and the degrees of freedom of movement of the connector arm assembly via a single locking actuator with one-handed operation. In embodiments, a control box computer system at the helmet communicates with various soldier-portable networks to send and receive information for display by a HUD device accessory.

Abstract: A detection device for detecting a fraud behavior of a customer at a point-of-sale (POS) terminal in a store, including a first interface circuit configured to receive a first image of the customer from a camera, a second interface circuit configured to receive a second image that is displayed on the POS terminal, and a processor configured to acquire one or more first images via the first interface circuit and determine one or more behaviors of the customer based on the acquired first images, acquire one or more second images via the second interface circuit and determine, based on the acquired second images, one or more operations that have been made by the customer on the POS terminal, and determine that one of the behaviors is fraudulent based on one or more of the operations that have been made by the customer before said one of the behaviors.

Abstract: The invention is based, in part, on a system and method designed to be able to easily and automatically scale up to millions of cameras and users. To do this, this discourse teaches use of modern cloud computing technology, including automated service provisioning, automated virtual machine migration services, RESTful API, and various firewall traversing methods to facilitate the scaling process. Moreover, the system and method described herein teaches scalable cloud solutions providing for higher though-put camera provisioning and event recognition. The network may segregate the retrieval server from the storage server, and by doing so, minimizing the load on any one server and improving network efficiency and scalability.

Abstract: Techniques are provided for determining confident data samples for machine learning (ML) models on unseen data. In one embodiment, a method is provided that comprises extracting, by a system comprising a processor, a feature vector for a data sample based on projection of the data sample onto a standard feature space. The method further comprises processing, by the system, the feature vector using an outlier detection model to determine whether the data sample is within a scope of a training dataset used to train a machine learning model, wherein the outlier detection model was trained using features extracted from the training dataset based on projection of data samples included in the training dataset onto the standard feature space.

Abstract: An apparatus comprises a camera and one or more sensors. The camera generally has a low power deactivated mode. The one or more sensors are generally remotely located with respect to the camera. The one or more sensors may be configured to communicate a signal to the camera in response to a trigger condition. The camera is generally configured to activate in response to receiving the signal from the one or more sensors.

Abstract: A method for contrast-based autofocus of an image capture device. The method includes obtaining sensor data representative of an image captured by the image capture device, wherein the sensor data comprises pixel values from respective sensor pixels of an image sensor of the image capture device. A subset of the pixel values is dynamically selected to generate selected sensor data representative of the subset of the pixel values. The selected sensor data is processed to generate contrast data representative of a contrast-based characteristic of at least a portion of the image. The contrast data is processed to determine a focus setting for the image capture device.

Abstract: Systems and methods are provided for presenting, within a page or user interface, a multi-layered video presentation that includes an interactive content portion presented over or otherwise in conjunction with a video player layer. Video content may be played in the video player layer while interactive selectable options associated with various items are presented in a separate layer in a time synchronized manner within the interactive content portion. User selection of an item may cause code associated with the interactive content portion to send an acquisition request for the item, and the interactive content portion may be updated to display a transaction confirmation associated with the item as the video content continues to play.

Abstract: One embodiment provides an apparatus comprising an input connector directly mateable with a media device to receive an analog signal. The apparatus further comprises an amplifier configured to amplify the analog signal, an analog-to-digital converter configured to convert the analog signal to a digital signal, a network interface configured to convert the digital signal to a networked compliant digital signal compatible with at least one network protocol, and an output connector mateable with a standard network cable that transmits the networked compliant digital signal.

Abstract: A method of electronic image color correction may include receiving a first frame including pixel data for a first color component for a plurality of pixels. The method may include generating saved pixel data for the first color component and receiving a second frame subsequent to the first frame. The second frame may include pixel data for the first color component for the plurality of pixels. The method may include generating substitute pixel data for the first color component in response to determining that, in the second frame, the pixel data for the first color component is triggered by a threshold value for a particular percentage of the plurality of pixels. The substitute pixel data may be generated based on the saved pixel data. The method may include displaying the second frame using the substitute pixel data for the first color component for the plurality of pixels.

Abstract: Various embodiments for providing removable, pluggable and disposable opto-electronic modules for illumination and microscopic imaging are provided, for use with portable display devices. Generally, various medical or industrial miniature microscopes can include one or more solid state or other compact electro-optic illuminating elements, electronic vision systems and means of scanning located thereon. Additionally, such opto-electronic modules may include illuminating optics, imaging optics, and/or image manipulation and processing elements. The illuminating elements may have different wavelengths and can be time-synchronized with an image sensor to illuminate an object for imaging or detecting purpose or other conditioning purpose. All control and power functions of such disposable microscope units can be made in the control unit that the disposable microscopes are plugged into.