A light field synthesis method and a light field synthesis system are provided. The light field synthesis method includes inputting light field information corresponding to a scene into a trained learning model. The light field information is a light field having a plurality of views. The light field synthesis method further includes configuring the trained learning model to generate a synthesized light field according to the input light field information. The synthesized light field has a plurality of new views other than the plurality of views. The trained learning model is obtained by performing a training process on a learning model, and the training process includes optimizing the learning model in a refocused image domain, so as to minimize refocused image errors.

A light field synthesis method and a light field synthesis system are provided. The light field synthesis method includes inputting light field information corresponding to a scene into a trained learning model. The light field information is a light field having a plurality of views. The light field synthesis method further includes configuring the trained learning model to generate a synthesized light field according to the input light field information. The synthesized light field has a plurality of new views other than the plurality of views. The trained learning model is obtained by performing a training process on a learning model, and the training process includes optimizing the learning model in a refocused image domain, so as to minimize refocused image errors.

A system and method for learning human activities from video demonstrations using video augmentation is disclosed. The method includes receiving original videos from one or more data sources. The method includes processing the received original videos using one or more video augmentation techniques to generate a set of augmented videos. Further, the method includes generating a set of training videos by combining the received original videos with the generated set of augmented videos. Also, the method includes generating a deep learning model for the received original videos based on the generated set of training videos. Further, the method includes learning the one or more human activities performed in the received original videos by deploying the generated deep learning model. The method includes outputting the learnt one or more human activities performed in the original videos.

A system and method for learning human activities from video demonstrations using video augmentation is disclosed. The method includes receiving original videos from one or more data sources. The method includes processing the received original videos using one or more video augmentation techniques to generate a set of augmented videos. Further, the method includes generating a set of training videos by combining the received original videos with the generated set of augmented videos. Also, the method includes generating a deep learning model for the received original videos based on the generated set of training videos. Further, the method includes learning the one or more human activities performed in the received original videos by deploying the generated deep learning model. The method includes outputting the learnt one or more human activities performed in the original videos.

A stored volumetric scene model of a real scene is generated from data defining digital images of a light field in a real scene containing different types of media. The digital images have been formed by a camera from opposingly directed poses and each digital image contains image data elements defined by stored data representing light field flux received by light sensing detectors in the camera. The digital images are processed by a scene reconstruction engine to form a digital volumetric scene model representing the real scene. The volumetric scene model (i) contains volumetric data elements defined by stored data representing one or more media characteristics and (ii) contains solid angle data elements defined by stored data representing the flux of the light field. Adjacent volumetric data elements form corridors, at least one of the volumetric data elements in at least one corridor represents media that is partially light transmissive. The constructed digital volumetric scene model data is stored in a digital data memory for subsequent uses and applications.

A stored volumetric scene model of a real scene is generated from data defining digital images of a light field in a real scene containing different types of media. The digital images have been formed by a camera from opposingly directed poses and each digital image contains image data elements defined by stored data representing light field flux received by light sensing detectors in the camera. The digital images are processed by a scene reconstruction engine to form a digital volumetric scene model representing the real scene. The volumetric scene model (i) contains volumetric data elements defined by stored data representing one or more media characteristics and (ii) contains solid angle data elements defined by stored data representing the flux of the light field. Adjacent volumetric data elements form corridors, at least one of the volumetric data elements in at least one corridor represents media that is partially light transmissive. The constructed digital volumetric scene model data is stored in a digital data memory for subsequent uses and applications.

[Problem] The present invention provides a stereoscopic image display device, stereoscopic image display method, and a program capable of displaying the display target with a constant size and a constant aspect ratio even when the distance between the stereoscopic image display device and the user changes. [Solution to Problem] The present invention provides a stereoscopic image display device comprising: a display part; an acquisition part configured to acquire an observation viewing distance that is a viewing distance from the display part to a user; and an adjustment part configured to adjust a display state of a display image based on a reference viewing distance and the observation viewing distance, wherein the display part is configured to display a stereoscopic image based on the display state.

In accordance with embodiments, a method of transmitting a video includes inter-view redundancy removing pictures for multiple viewing positions; packing inter-view redundancy removed pictures; and/or encoding the packed pictures and/or signaling information including center view generation information, pre-generation information, view synthesis recommendation information and/or reference view information. In accordance with embodiments, a method of receiving a video includes decoding a bitstream of the video based on viewing position and/or viewport information, unpacking pictures in the decoded bitstream; view regenerating the unpacked pictures; and/or view synthesizing the view regenerated pictures.

In accordance with embodiments, a method of transmitting a video includes inter-view redundancy removing pictures for multiple viewing positions; packing inter-view redundancy removed pictures; and/or encoding the packed pictures and/or signaling information including center view generation information, pre-generation information, view synthesis recommendation information and/or reference view information. In accordance with embodiments, a method of receiving a video includes decoding a bitstream of the video based on viewing position and/or viewport information, unpacking pictures in the decoded bitstream; view regenerating the unpacked pictures; and/or view synthesizing the view regenerated pictures.

An image display device includes a processor that sets a location of a virtual image plane on which a virtual image is formed according to depth information included in first image data and generates second image data obtained by correcting the first image data based on the set location of the virtual image plane; an image forming optical system including a display element configured to modulate light to form a display image according to the second image data and a light transfer unit that forms the virtual image on the virtual image plane, the virtual image corresponding to the display image formed by the display element, the light transfer unit comprising a focusing member; and a drive unit that drives the image forming optical system to adjust the location of the virtual image plane.