123 results about "Visual Disparity" patented technology

A three-dimensional display is provided which can produce a stereoscopic image with a natural stereoscopic depth even on different screen sizes. A stereoscopic video signal generation circuit, which supplies a stereoscopic video signal to the three-dimensional display that forms a stereoscopic image by taking advantage of binocular disparity parallax, comprises: an information retrieving means for retrieving video information on the stereoscopic image and display information on the three-dimensional display; and an offset setting means for offsetting a left-eye image and a right-eye image relative to each other according to the video information and the display information to adjust the stereoscopic depth of the image displayed.

A mobile terminal and a method for controlling the operation of the mobile terminal are provided. In the method, a stereoscopic 3D image that utilizes binocular disparity between a left-eye image and a right-eye image is displayed on a display and a stereoscopic 3D pointer is displayed in the stereoscopic 3D image when a user's hand is detected according to analysis of visual data acquired by a camera. The position of the stereoscopic 3D pointer is moved according to movement of the hand when the movement of the hand is detected according to analysis of the visual data. According to this method, the operation of the mobile terminal can be controlled by the user's hand movement or gesture.

The invention discloses a barrier and travelable area detection method based on a vehicle-mounted binocular camera. The barrier and travelable area detection method comprises the following steps of S1, taking an image through a binocular camera; S2, preprocessing the image; S3, calculating the final matching cost of each pixel point in the image; S4, calculating through the final matching cost to obtain a v-disparity map; S5, determining the horizon line and the road scope by means of the v-disparity map; S6, calculating to obtain a barrier-road intersecting line in the road scope; S7, determining the scope of the current lane and the outermost lane; S8, calculating to obtain the height of the barrier on the part above the barrier-road intersecting line; S9, determining the image areas which belong to the same barrier; and S10, outputting the road area information and barrier information. Under the premise that disparity precision is not reduced, unnecessary binocular disparity map calculation is avoided, and therefore the real-time performance of the algorithm is greatly improved, and the barrier and travelable area detection method based on the vehicle-mounted binocular camera is suitable for fields of driving warning, automatic driving and the like.

An image processing apparatus and method may estimate binocular disparity maps of middle views from among a plurality of views through use of images of the plurality of views. The image processing apparatus may detect a moving object from the middle views based on the binocular disparity maps of the frames. Pixels in the middle views may be separated into dynamic pixels and static pixels through detection of the moving object. The image processing apparatus may apply bundle optimization and a local three-dimensional (3D) line model-based temporal optimization to the middle views so as to enhance binocular disparity values of the static pixels and dynamic pixels.

A stereo image display apparatus is disclosed, in which the influence of display area edge portions on the viewing of stereo image is taken into considerations to enable display of stereo images without spoiling the viewer's sense just like the viewer is actually on the site of the image scene. The display of a left eye and a right eye image is controlled to have a binocular parallax substantially fixed in effect. (Specifically, image shifters 32L and 32R shift the left eye and right eye images according to a parallax signal.) Edge portions of a left eye and a right eye image display area are shaded off in effect and made obscure by a left eye and a right eye eyepiece optical system 12L and 12R.

A method for controlling a mobile terminal image includes providing a first image and a second image via a controller on the mobile terminal, the first and second images reflecting a binocular disparity to form a three dimensional image, identifying an editing target from the three dimensional image, editing a first image of the identified editing target, and applying the edited first image and a second image corresponding to the edited first image to the three dimensional image.

The invention provides a double-camera color synchronization method and device, and a terminal. The method comprises the steps that the double cameras obtain Bayer images; the white balance gain parameters of the Bayer images are calculated, and current environment light source color temperature is obtained; a color synchronization coefficient is obtained by employing the calculated color temperature according to a color synchronization coefficient mapping table; a digital signal processor applies the color synchronization coefficient to a next Bayer image photographed by a non-reference camera. According to the method, the device and the terminal, the image data is based on the Bayer image data; compared with 8-bit images processed by an image signal processor ISP, 10-bit Bayer images have richer data information, and therefore, the effect of an image synthesized by employing the Bayer images is better. According to the method, the device and the terminal, the original data color consistency of the double cameras is calibrated, the vision difference resulting from color inconsistency is improved, the synthesizing effect of the double cameras is improved, and the user experience is improved.

A method of controlling a mobile terminal, and which includes receiving a selection signal from an input unit setting a 3D attribute to at least one item among a plurality of items to be displayed on a display of the mobile terminal; and turning on a switching panel unit positioned in front of the display, via a controller controlling the switching panel unit, when the at least one item is displayed on the display of the mobile terminal. Further, the switching panel unit displays left and right eye images of the at least one item such that the at least one item is viewed as a 3D image based on binocular disparity.

A tile-based rendering method and a tile-based rendering apparatus are provided. The tile-based rendering method involves performing tile binning on a first image and a second image having binocular disparity using a plurality of graphic processors based on a draw command, determining a tile of the second image having a highest similarity to a tile of the first image based on a result of the tile binning, determining a tile rendering order for each tile of the first image and the second image, based on a result of the determining of the tile of the second image having a highest similarity to the tile of the first image, and performing tile rendering, according to the determined tile rendering order.

The invention belongs to the image processing field and relates to a stereo image quality evaluation method based on GSSIM. The method includes the steps: (1) for left image and right image, gradient structure similarity values are respectively solved, the average of the two values is solved, so as to obtain stereo image quality evaluation value QE; (2) the following method is adopted to carrying out image stereoscopic perception objective evaluation: absolute difference image of the original image and processed image is calculated, Mu1 and Mu2 of the absolute difference image are solved; Sobel operator is used for solving the gradient amplitude image of the absolute difference image; filtering is carried out on the absolute difference image of the original image, and original image binocular parallax distribution situation is calculated; Dl(x, y), Dcg(x, y) and Dsg(x, y) values at binocular parallax position are solved; DSSIM value at the binocular parallax position is calculated; and image gradient structure similarity namely image stereoscopic perception objective evaluation value DE is calculated. The invention can be well applied to stereo image quality evaluation, and correlation of objective evaluation result and subjective evaluation result is strong.

The application provides a generation method, apparatus and terminal of a disparity map, and relates to the technical field of auxiliary driving. The method includes: obtaining a binocular source image acquired by a binocular camera; performing abbreviation processing on the binocular source image to obtain a thumbnail of the binocular source image; performing stereo vision matching on the thumbnail to obtain a thumbnail disparity map; determining a target area corresponding to a physical detection area in the thumbnail disparity map in a real space; mapping the target area to an area in the binocular source image as an obstacle detection area; and performing stereo vision matching on the obstacle detection area in the binocular source image to obtain a binocular disparity map. By adoptingthe method, disparity estimation of pixel points having significance for obstacle detection can be performed in a targeted manner, calculating resources can be effectively utilized, the efficiency ofa stereo matching algorithm is improved, and the efficiency for subsequent obstacle detection based on the disparity map is further improved.

The invention provides a binocular disparity estimation method based on a cascaded geometric context neural network. The invention designs a new cascaded convolution neural network Cascaded GCNet (CGCNet). The network mainly adopts the improved GCNet and combines the 3d convolution and the original 2d convolution operation to obtain a better disparity map characteristic expression, which is beneficial to the subsequent network training. RefineNet is used to optimize the rough parallax map of GCnet network, and the precision of parallax map is improved by iterative refinement. In the process ofRefineNet optimization, difficult mining is used to make the network model focus on learning rare samples, so as to improve the disparity estimation accuracy of the network for image pairs with different complexity.

The invention discloses an unsupervised monocular depth estimation method based on a generative adversarial network, and the method comprises the following steps: 1, obtaining a left and right image pair with strict time synchronization through a binocular camera, building a binocular color image data set, and correcting a binocular color image; 2, establishing an unsupervised generative adversarial network model, inputting the corrected binocular color image into the network, and performing training and iterative regression on the network model; 3, inputting the monocular color image into thetrained network model to generate a disparity map corresponding to the monocular color image; and 4, converting the disparity map into depth information through a binocular disparity depth conversionformula, and synthesizing a depth map. According to the depth estimation method provided by the invention, the monocular color image is converted into the depth map containing the depth information by using the unsupervised network model, and complex real depth data is not needed.

The invention discloses a method and device for detecting a road obstacle. The method comprises: acquiring road images through a binocular device and processing the road images to obtain a disparity map of the road images; carrying out pavement-information-based image segmentation to obtain a binary image with an obstacle region as a foreground and other regions as a background; carrying out morphological operation on the binary image to obtain a to-be-detected region template of a suspected obstacle; carrying out information fusion of the to-be-detected region template and a monocular image being one of the road images to obtain a grayscale image of the to-be-detected region; and inputting the grayscale image into a preset obstacle detection model to carry out obstacle determination on the road images and determining the suspected obstacle based on a determination result. According to the invention, on the basis of fusion of respective advantages of binocular disparity information andmonocular image information, machine-learning-based target recognition and target classification are carried out to obtain a quick and accurate obstacle detection result.

The invention discloses a binocular parallax estimation method based on three-dimensional convolution. Includes the following steps: calibrating the binocular camera by using the binocular fixing method to obtain a binocular correction map; The binocular image to be estimated is corrected to obtain the corrected binocular image; The corrected binocular image is sent into the preset two-dimensionalconvolution neural network to obtain the feature map after feature transformation. The feature map of the current frame and the feature map of the previous multi-frame images are stitched and sent tothe three-dimensional convolution neural network to obtain the feature map of the multi-frame images. The characteristic images of multi-frame images are transposed and convolution, and then transformed back to pixel domain to obtain disparity estimation map. Compared with the prior binocular parallax estimation method based on the convolution neural network, the invention estimates the binocularparallax map of the current frame by extracting the information on the time dimension through the three-dimensional convolution and combining the binocular information of the current frame and the previous multiple frames. Compared with the original method, this method improves the accuracy and the coherence between the two frames.

The invention discloses a depth estimation method based on binocular vision and laser radar fusion. The method comprises the following steps: registering data acquired by a laser radar and a binocularcamera through joint calibration; obtaining a laser radar disparity map according to a joint calibration result; obtaining a binocular disparity map through a binocular stereo matching algorithm, performing confidence analysis on the binocular disparity map, removing points with low confidence in the binocular disparity map, and obtaining the binocular disparity map after confidence processing; performing feature extraction and fusion on the obtained laser radar disparity map and the binocular disparity map after confidence processing; carrying out further feature extraction and parallax regression through a cascade hourglass structure; adopting relay supervision to utilize output of front and back cascaded hourglass structures; and outputting an accurate and dense disparity map after fusion. According to the method, a more effective network structure is designed, the features of the laser radar disparity map and the binocular disparity map are better extracted and fused, and a more accurate disparity map is obtained.

Light that is incident through a pair of object lenses of digital binocular glasses is photographed by a pair of image pickup devices to obtain a pair of images (stereoscopic image) generating a difference corresponding to a parallax of both eyes. A geometric difference between image structures corresponding to a parallax of both eyes is recognized in the pair of images subjected to various corrections and stored in memories. Thereafter, a noise reduction process is performed, which reduces a difference (for example, a difference between random noises superposed on the pair of images by the image pickup devices) other than the recognized geometric difference between the image structures. The images subjected to the noise reduction process are displayed on a pair of display devices, and are visually recognized (stereoscopically viewed) by the user through eyepiece lenses.

An apparatus is provided for converting two-dimensional video frames to stereoscopic video frames, comprising a comparing unit, a temporal transform operator and a spatial transform operator. The comparing unit receives a sequence of two-dimensional video frames and provides a comparison result indicative of temporal difference data for the sequence of two-dimensional video frames. The temporal transform operator is coupled to the comparing unit for modifying the sequence of two-dimensional video frames according to the comparison result. The spatial transform operator is coupled to the comparing unit and the temporal transform operator for generating a stereoscopic frame pair according to each frame of the modified video frames. The stereoscopic frame pair comprises a left-eye video frame and a right-eye video frame for providing visual differences perceived by human eyes.

A mobile terminal and a method of controlling the operation of the mobile terminal are provided. The method includes dividing each of first and second images into a plurality of blocks, the first and second images capable of realizing a three-dimensional (3D) image using binocular disparity; searching the first and second images for pairs of matching blocks and calculating depth information of each of the pairs of matching blocks based on a difference in position between each of the pairs of matching blocks; and calculating stereoscopic information of the 3D image based on the depth information of each of the pairs of matching blocks. Therefore, it is possible to effectively provide stereoscopic information of a 3D image for various purposes.

The invention discloses an image defogging method for haze concentration of an adaptive neural network based on end-to-end. The method comprises the following steps: constructing an image defogging model; acquiring foggy image data; using a feature enhancement module in the image defogging model for cascading the feature map with images recovered by different paths, combining fuzzy images with different dense haze degrees to help the network sense image haze concentration in a self-adaptive mode; reconstructing the features after function enhancement into a clear fog-free image through a multi-scale feature attention module; calculating mean square errors and perception losses of the restored images and the corresponding clear images, and updating an image defogging model; wherein the meansquare errors guide the image defogging model to learn the content of the clear images, the perception loss is used for quantizing the visual difference between the recovered images and the corresponding clear images, and the two loss functions cooperatively optimize the defogging model. According to the technical scheme, effective defogging processing is carried out on the actually shot fog image, a high-quality image is recovered, and the practicability is good.

An apparatus and a method for generating Three Dimensional (3D) contents in an electronic device are provided. The method includes extracting data having geometric information, generating two images having binocular disparity using the geometric information of the extracted data, and outputting the generated two images to a display unit. The generating of the two images having the binocular disparity includes rendering a first image using the geometric information of the extracted data, and generating a second image using depth information of an object in the first image.

A depth estimating apparatus operated by at least one processor includes: a database which stores a photographed first color image, a training thermal image geometrically aligned with the first color image, and a second color image simultaneously photographed with the first color image as a training image set; and a training apparatus which trains a neural network in an unsupervised manner to output a chromaticity image and a binocular disparity image from the training thermal image. The training apparatus generates an estimated first color image from the second color image, the chromaticity image, and the binocular disparity image, and trains the neural network to minimize a difference between the estimated first color image and the photographed first color image.

A stereomicroscope includes a microscope main body and a camera for assistant. The microscope main body includes therein an objective optical system, a zoom optical system and an eyepiece optical system. The camera for assistant is detachably mounted to the microscope main body. The microscope main body provides to a main operator a pair of optical images including a certain binocular disparity as viewed from an observation direction of the main operator. The camera for assistant provides to an assistant a pair of electron images including a certain binocular disparity as viewed from an observation direction of the assistant. The observation directions of the main operator and the assistant cross at a certain angle. The camera for assistant outputs the pair of electron images into a stereo viewer which is separated from the microscope main body.

The present application relates to a device and method for processing digital broadcast signals comprising three-dimensional images. According to one embodiment of the present invention, the digital broadcast signal processing method comprises the steps of: encoding a video stream of a two dimensional reference image; encoding a video stream of an additional image, for generating a binocular-disparity three-dimensional image together with the encoded two dimensional reference image; generating signaling information such that the video stream of the encoded additional image is processed only in a three-dimensional image signal receiver; generating a first broadcast signal comprising the video stream of the encoded reference image; generating a second broadcast signal comprising the video stream of the encoded additional image and the signaling information; and respectively sending the first broadcast signal and the second broadcast signal over different channels.

A method for rendering an object in a 3D graphic terminal includes constructing camera coordinates, based on vertex information of objects existing in a 3D space, and selecting one object in a left frustum and a right frustum, based on the constructed camera coordinates, wherein the left frustum is defined centered on a left virtual camera viewpoint, and the right frustum is defined centered on a right virtual camera viewpoint. The method further includes determining a binocular disparity by projecting vertexes of the selected object in the left frustum and the right frustum, and adjusting frustum parameters of the left virtual camera and the right virtual camera when the determined binocular disparity is greater than an allowable binocular disparity.

The invention provides a shell assembly. The assembly comprises a transparent shell and an electrochromic layer, the transparent shell comprises a first area and a second area adjacent to the first area. The first area comprises a window area and a non-window area adjacent to the window area; the electrochromic layer is arranged on the inner surface of the transparent shell, the orthographic projection of the electrochromic layer on the transparent shell completely covers the window area, the electrochromic layer has a colored state and a transparent state, when the electrochromic layer is inthe colored state, the color difference value of the window area and the non-window area is smaller than or equal to 1, and the second area and the non-window area have visual difference. The electrochromic layer is arranged at the position corresponding to the window area, so that the color of the window area is consistent with that of the non-window area when the electrochromic layer is in a colored state, the visual effect of local integration of the shell assembly is achieved, meanwhile, the second area and the non-window area have visual difference, and the appearance expressive force ofthe shell assembly is further improved. The invention further provides a preparation method of the shell assembly and electronic equipment.

The present invention relates to a stereoscopic presentation system, and more specifically to a stereoscopic presentation system which can show documents stereoscopically by compiling text, pictures, diagrams or moving images, which constitute two-dimensional objects, into 3-dimensional objects which are output on a stereoscopic image display unit. A stereoscopic presentation system according to one embodiment of the present invention comprises: a two-dimensional object compiling unit for compiling text, pictures, diagrams or moving images, which constitute two-dimensional objects; a three-dimensional object compiling unit for compiling the input two-dimensional objects into three-dimensional objects having a set binocular disparity; a three-dimensional object storage unit for storing the three-dimensional objects; and a stereoscopic image display unit for outputting the three-dimensional objects stored in the three-dimensional object storage unit.

An external scene image captured by an external scene imaging electronic camera attached to a head mounted display (HMD) is projected and displayed onto an image display screen arranged in front of the eyes of the user as a virtual image with a suitable viewing distance corresponding to the visual acuity of the user. At this time, for each object image presented in the virtual image of the external scene image, the virtual image is processed and formatted to add a predetermined degree of binocular disparity and image blur to the virtual image projected and displayed on the right and the left image display screen on the basis of a predetermined converted distance calculated from the real distance of each object. Thus, the user is given a sense of a realistic perspective for the virtual image of the external scene, free of the discomfort or unease.

A parallel axis stereoscopic camera comprising: a camera unit which includes left and right image sensors each of which has a higher resolution than an output image, the camera unit outputting RGB data having the same resolution as the output image; a vergence controller which performs an electronic control for eliminating a binocular disparity of an object by changing a read-out starting point in the horizontal direction, of at least one of the left and right image sensors; an image processor including a left image processing unit which processes an image of left RGB data to output a left luminance / chrominance signal and a right image processing unit which processes an image of right RGB data to outputs a right luminance / chrominance signal under the control of the vergence controller; and a stereoscopic image synthesizer which synthesizes the left and right luminance / chrominance signals to produce a stereoscopic image.