A stereoscopic imaging apparatus and platform are disclosed. An example stereoscopic imaging apparatus includes a main objective assembly and left and right lens sets defining respective parallel left and right optical paths from light that is received from the main objective assembly of a target surgical site. Each of the left and right lens sets includes a front lens, first and second zoom lenses configured to be movable along the optical path, and a lens barrel configured to receive the light from the second zoom lens. The example stereoscopic imaging apparatus also includes left and right image sensors configured to convert the light after passing through the lens barrel into image data that is indicative of the received light. The example stereoscopic visualization camera further includes a processor configured to convert the image data into stereoscopic video signals or video data for display on a display monitor.

A method of synthesizing intermediate views of a light field includes selecting a configuration of specific input views of a light field collected by a light field acquiring device, specifying coordinates of intermediate views to be synthesized and inputting the specified coordinates to a neural network, and synthesizing intermediate views based on a scene disparity, a selected configuration of the specific input views, and the specified coordinates of the intermediate views, using a neural network.

A method of synthesizing intermediate views of a light field includes selecting a configuration of specific input views of a light field collected by a light field acquiring device, specifying coordinates of intermediate views to be synthesized and inputting the specified coordinates to a neural network, and synthesizing intermediate views based on a scene disparity, a selected configuration of the specific input views, and the specified coordinates of the intermediate views, using a neural network.

There is provided an image processing apparatus comprising: a processor; and a memory storing a program. When the program is executed by the processor, the program causes the image processing apparatus to: obtain a first RAW image including a region of a first circular fisheye image; and develop the first RAW image. A pixel outside the region of the first circular fisheye image in the first RAW image is not developed.

There is provided an image processing apparatus comprising: a processor; and a memory storing a program. When the program is executed by the processor, the program causes the image processing apparatus to: obtain a first RAW image including a region of a first circular fisheye image; and develop the first RAW image. A pixel outside the region of the first circular fisheye image in the first RAW image is not developed.

A video transmission method that includes predicting, from a texture picture or a depth picture of an anchor viewing position, a picture for a target viewing position on the basis of target viewing position information and processing a prediction error with respect to the predicted picture on the basis of a source picture of the target viewing position. An error-prone region map is generated on the basis of the predicted picture and the source picture. The video transmission method also includes patch packing the prediction error-processed picture on the basis of the error-prone region map and encoding the packed patch on the basis of the texture picture or the depth picture of the anchor viewing position.

This disclosure relates to techniques for generating robust depth estimations for captured images using semantic segmentation. Semantic segmentation may be defined as a process of creating a mask over an image, wherein pixels are segmented into a predefined set of semantic classes. Such segmentations may be binary (e.g., a ‘person pixel’ or a ‘non-person pixel’) or multi-class (e.g., a pixel may be labelled as: ‘person,’ ‘dog,’ ‘cat,’ etc.). As semantic segmentation techniques grow in accuracy and adoption, it is becoming increasingly important to develop methods of utilizing such segmentations and developing flexible techniques for integrating segmentation information into existing computer vision applications, such as depth and/or disparity estimation, to yield improved results in a wide range of image capture scenarios. In some embodiments, an optimization framework may be employed to optimize a camera device's initial scene depth/disparity estimates that employs both semantic segmentation and color regularization in a robust fashion.

This disclosure relates to techniques for generating robust depth estimations for captured images using semantic segmentation. Semantic segmentation may be defined as a process of creating a mask over an image, wherein pixels are segmented into a predefined set of semantic classes. Such segmentations may be binary (e.g., a ‘person pixel’ or a ‘non-person pixel’) or multi-class (e.g., a pixel may be labelled as: ‘person,’ ‘dog,’ ‘cat,’ etc.). As semantic segmentation techniques grow in accuracy and adoption, it is becoming increasingly important to develop methods of utilizing such segmentations and developing flexible techniques for integrating segmentation information into existing computer vision applications, such as depth and/or disparity estimation, to yield improved results in a wide range of image capture scenarios. In some embodiments, an optimization framework may be employed to optimize a camera device's initial scene depth/disparity estimates that employs both semantic segmentation and color regularization in a robust fashion.

Systems, methods, and non-transitory computer-readable medium storing instructions that, when executed, causes a processor to perform operations to display a three-dimensional (3D) space. The methods may include, with an imaging device, capturing a first series of frames as the imaging device travels from a first location to a second location within a space, and capturing a second series of frames as the imaging device travels from the second location to the first location. The method may also include determining a first segment in the first series of frames that matches a second segment in the second series of frames to create a segmentation dataset, generating video clip data based on the segmentation dataset, the video clip data defining a series of video clips, and displaying the series of video clips.

Systems, methods, and non-transitory computer-readable medium storing instructions that, when executed, causes a processor to perform operations to display a three-dimensional (3D) space. The methods may include, with an imaging device, capturing a first series of frames as the imaging device travels from a first location to a second location within a space, and capturing a second series of frames as the imaging device travels from the second location to the first location. The method may also include determining a first segment in the first series of frames that matches a second segment in the second series of frames to create a segmentation dataset, generating video clip data based on the segmentation dataset, the video clip data defining a series of video clips, and displaying the series of video clips.