A machine-learned image compression model includes a first encoder configured to generate a first image code based at least in part on first image data. The first encoder includes a first series of convolutional layers configured to generate a first series of respective feature maps based at least in part on the first image. A second encoder is configured to generate a second image code based at least in part on second image data and includes a second series of convolutional layers configured to generate a second series of respective feature maps based at least in part on the second image and disparity-warped feature data. Respective parametric skip functions associated convolutional layers of the second series are configured to generate disparity-warped feature data based at least in part on disparity associated with the first series of respective feature maps and the second series of respective feature maps.

Thermal imaging camera images are obtained from a thermal imaging camera that rotates through a plurality of stop positions. The camera captures images at a constant frame rate and at least some of the images correspond to stop positions. Thermal imaging camera images that correspond to a stop position are retained, while images that do not correspond to a stop position are discarded. Retained images are sent in a video stream to a video processor. The video stream is separated into individual thermal imaging camera images and stored for corresponding virtual camera devices that correspond to specific stop positions. In addition, the position of the camera and individual pixels of images are both correlated to geographical location data, and depth values for the pixels are determined based on the geographical data.

Dense field imagers are disclosed which are configured to provide combined, aggregated, fused, and/or stitched light field data for a scene. A dense field imager can include a plurality of imaging elements configured to be joined into image blocks or facets that each provides light field data about a scene. The dense field imager can include a plurality of facets in a fixed or modular fashion such that the dense field imager is configured to combine, aggregate, fuse and/or stitch light field data from the plurality of facets. The facets can be mounted such that one or more facets are non-coplanar with other facets. The facets can be configured to provide a representation of the light field with overlapping fields of view. Accordingly, the dense field imager can provide dense field data over a field of view covered by the plurality of facets.

Disclosed are an electronic apparatus that achieves both the convenience of touch operation and accurate position specification, and a method for controlling the same. When an area that meets a predetermined condition has been detected in an image, the electronic apparatus executes a function corresponding to a position where a touch has been detected by touch detection means. When the area that meets the predetermined condition has not been detected or area detection is not carried out, the electronic apparatus carries out a process corresponding to an operation for moving a touched position rather than carrying out processing corresponding to the position where the touch has been detected by the touch detection means.

A seamless lens cover, and methods of forming such a seamless lens cover. The cap structure that covers a camera of a rotating panoramic camera system includes a seamless lens cover through which images are obtained by the camera. The cap structure may be injection molded at an initial lens cover thickness, and then a portion of the as molded initial lens cover thickness may be removed (e.g., by machining away) to achieve the final desired thickness. By such a method, the lens cover may be injection molded at thicknesses suitable for injection molding (e.g., about 0.06 to about 0.1 inch), after which most of the thickness may be machined away, to provide a seamless lens cover having a thickness of less than about 0.015 inch, exhibiting at least 60% transmittance to the thermal spectrum, no lensing characteristics, and no curvature effect.

An optical-scanning observation device including: a laser output unit that outputs a laser beam; a light scanner that radiates the laser beam on an object while scanning; a light detector that detects reflected light of the laser beam; and a controller that sets a detection delay time on the basis of a propagation delay time from when the laser beam is output from the laser output unit to when the reflected light is detected by the light detector and that controls the laser output unit and the light detector such that the light detector detects the reflected light after the detection delay time has elapsed since the laser beam is output from the laser output unit.

Thermal imaging camera images are obtained from a thermal imaging camera that rotates through a plurality of stop positions. The camera captures images at a constant frame rate and at least some of the images correspond to stop positions. Thermal imaging camera images that correspond to a stop position are retained, while images that do not correspond to a stop position are discarded. Retained images are sent in a video stream to a video processor. The video stream is separated into individual thermal imaging camera images and stored for corresponding virtual camera devices that correspond to specific stop positions. In addition, the position of the camera and individual pixels of images are both correlated to geographical location data, and depth values for the pixels are determined based on the geographical data.

In some implementations a mountable afocal adaptor for a camera includes an attachment plate affixed to the front of a camera housing of a camera and an afocal optical module. The afocal optical module is configured to convert a received optical image to a converted optical image for use by the camera. The afocal adaptor also includes a rotation mechanism with a rotation ring and a wave spring clamp, the rotation ring affixed to the attachment plate and the wave spring clamp affixed to the afocal optical module. The rotation ring and wave spring clamp are mounted relative to each other to rotate around a rotational axis aligned with an axis of an optical path formed between the afocal optical module and the attachment plate and aligned with a view axis of the camera. The rotation mechanism also permits rotation of the camera housing around the rotational axis relative to the afocal optical module.