The invention relates to a stereoscopic reproduction system without glasses, comprising an image reproduction device using transparency acting as means for illuminating multiple viewers with polarized light, two image capture elements separated from one another, containing polarizing filters that allow each of them to capture two images, one with the reflection of the light in the corneas and the other one eliminating said reflection, processing means for obtaining the pixel-by-pixel subtraction of those two images, the location of each eye in each image being obtained as the geometric center of the reflections in each cornea and the parallax of each eye between the two capture elements. With this data, the processing means generate as many luminous rectangles on a screen which backlights an array of converging lenses, which in turn focuses each of these luminous rectangles on the eye of each viewer after going through an image reproduction screen using transparency in which two images corresponding to the right eye and to the left eye are reproduced in synchronization with illumination of the luminous rectangles corresponding to each eye.

An information processing device according to the present technology includes a selection processing unit that selects data to be used to generate a free viewpoint image according to an importance related to at least one of an event or a viewpoint, as setting a plurality of pieces of captured image data obtained by imaging the event from the plurality of viewpoints and processing data obtained by executing at least processing related to three-dimensional information generation of a subject on the captured image data as selection target data.

A method for detecting three-dimensional (3D) pseudoscopic images and a display device for detecting 3D pseudoscopic images are provided. The method includes extracting corresponding feature points in a first view and corresponding feature points in a second view, wherein the first view and the second view form a current 3D image; calculating an average coordinate value of the feature points in the first view and an average coordinate value of the feature points in the second view; based on the average coordinate value of the feature points in the first view and the average coordinate value of the feature points in the second view, determining whether the current 3D image is pseudoscopic or not; and processing the current 3D image when it is determined that the current 3D image is pseudoscopic.

A method is disclosed for detecting interaction between two or more participants in a meeting, which includes capturing at least one three-dimensional stream of data on the two or more participants; extracting a time-series of skeletal data from the at least one three-dimensional stream of data on the two or more participants; classifying the time-series of skeletal data for each of the two or more participants based on a plurality of body position classifiers; and calculating an engagement score for each of the two or more participants. In addition, a method is disclosed for improving a group interaction in a meeting, which includes calculating, for each of the two or more participants, an individual engagement state based on attitudes of the participant, wherein the individual engagement state is an engagement state of the participant to the meeting including an engaged state and a disengaged state.

An image recognition approach employs both computer generated and manual image reviews to generate image tags characterizing an image. The computer generated and manual image reviews can be performed sequentially or in parallel. The generated image tags may be provided to a requester in real-time, be used to select an advertisement, and/or be used as the basis of an internet search. In some embodiments generated image tags are used as a basis for an upgraded image review. A confidence of a computer generated image review may be used to determine whether or not to perform a manual image review.

An image pickup apparatus 12 includes a first camera 22 and a second camera 24. The cameras pick up an image of a target at the same timing and at the same frame rate from left and right positions spaced by a known distance from each other. Picked up frame images are converted into image data of a predetermined plurality of resolutions. An input information acquisition section 26 of an information processing apparatus 14 acquires an instruction input from a user. A position information production section 28 approximately estimates a region of the target or a region which involves some movement on an image of a low resolution and a wide range from within stereo image data as a target region and carries out stereo matching only in regard to the region on an image of a high resolution to specify a three-dimensional position of the target. An output information production section 32 carries out a necessary process based on the position of the target to produce output information. A communication section 30 carries out request for and acquisition of image data to and from the image pickup apparatus 12.

A method of processing a stereoscopic video includes determining whether a current frame of a stereoscopic video is a video segment boundary frame; determining whether an image error is included in the current frame when the current frame is the video segment boundary frame; and processing the current frame by removing, from the current frame, a post inserted object (PIO) included in the current frame when the image error is included in the current frame.

The present invention discloses a non-feature extraction dense SFM three-dimensional reconstruction method, comprising: inputting n images about a certain scenario, n2; establishing a world coordinate system consistent with a certain camera coordinate system; constructing an objective function similar to optical flow estimation by taking a depth of a three-dimensional scenario and a camera projection matrix as variables; employing a from coarse to fine pyramid method; designing an iterative algorithm to optimize the objective function; outputting depth representing the three-dimensional information of the scenario and a camera projection matrix representing relative location and pose information of the camera; and realizing dense projective, similarity or Euclidean reconstruction according to the depth representing the three-dimensional information of the scenario. The present invention can accomplish dense SFM three-dimensional reconstruction with one step. Since estimation of dense three-dimensional information is achieved by one-step optimization, an optimal solution or at least local optimal solution can be obtained by using the objective function as an index, it is significantly improved over an existing method and has been preliminarily verified by experiments.

There is provided an image processing apparatus including a stereo matching unit configured to obtain right and left disparity images by using stereo matching, based on a pair of images captured by right and left cameras, respectively, a filter processing unit configured to perform filter processing on the disparity images, and a first merging unit configured to make a comparison, in the disparity images that have undergone the filter processing, between disparity values at mutually corresponding positions in the right and left disparity images and to merge the disparity values of the right and left disparity images based on a comparison result.

Techniques to determine a search range for a stereo based matching pyramid. A first disparity estimation value for a first level in a stereo based matching pyramid based on an image may be received. A search range for a second level may be determined using the first disparity estimation value. The search range based on a pyramid level of a second level may be increased. The search range may be increased based on a pyramid level of the second level. A second disparity estimation value may be selected from the search area for the second level. A depth map for the second level may be determined based on the second disparity estimation value. Other embodiments are described and claimed.