Alcatel Lucent

Donny G Tytgat

Sammy Lievens

Maarten Aerts

Jean-Francois Macq

Erwin Six

Christoph Stevens

Erwin Alfons Constant Six

A method for generating a set of respective transformation maps for a set of respective 2D images from a same object and using a parameter-based transformation model, comprises the steps of &#x2014;receiving said set of respective 2D images and said parameter-based transformation model &#x2014;detecting matching regions across several pairs of the 2D images, based on the set of 2D images and 3D information of said object, &#x2014;identifying respective interdependencies of the matching regions over the 2D images, &#x2014;optimizing the parameters of the parameter-based transformation model over the matching regions of all images as well as over the non-matching regions in all images.

Method and device for image processing

The present invention relates to a method, system and related devices for navigating between navigation points of a 3-Dimensional space where said 3-dimensional space comprises a plurality of navigation points. The method according to the present invention comprises the steps of detecting a current viewpoint rotation angle, further determining a degree of agreement of said current viewpoint rotation angle with a navigation point of said plurality of navigation points and subsequently activating a timing mechanism of the viewpoint activation, said timing mechanism comprising a delay for activating said timing mechanism of the viewpoint activation and a duration of an activation period of said timing mechanism at determining a certain degree of agreement of said current viewpoint rotation angle with a navigation point of said plurality of navigation points. The method further comprises the step of determining a degree of agreement of a portion of a viewpoint rotation angle trace with a viewpoint rotation model and adapting said at least one of said delay for activating said timing mechanism of said viewpoint activation and said duration of said activation period of said timing mechanism based on at least one of said degree of agreement of said current viewpoint rotation angle with a navigation point of said plurality of navigation points and said degree of agreement of said portion of said viewpoint rotation angle trace with said viewpoint rotation model.

Method for navigating between navigation points of a 3-dimensional space, a related system and a related device

Video decoder adapted for decoding video based on decoder parameters selected from variable decoder parameters, the decoder comprising an estimator adapted to estimate user viewing experience based on sensor data and comprising a constraint analyzer adapted to analyze constraints when using the decoder parameters, the video decoder further comprising a selector adapted to select said decoder parameters from the variable decoder parameters, wherein the selector is coupled to the estimator and the constraint analyzer.

Video decoding and encoding system

The present invention relates to a method, system and related devices for annotating an object in a multimedia asset, where the multimedia asset comprises at least one image. The method comprises the steps of first determining a category of the object in an image of the multimedia asset by means of classifying the object and subsequently selecting based on the category of the object an appropriate object detector model from at least one object detector model associated with said category. Thereafter a location of the object in the image of the multimedia asset is determined based on the object detector model that is selected, where the object detector model corresponds to the object which step is followed by segmenting the object detected in the image of the multimedia asset based on the location of the object within the said image and optionally based on the object detector. Finally, the object within the image being segmented is annotated with a hyperlink.

Method for annotating an object in a multimedia asset

A method and device for determining at least one set of matching attributes between a plurality of images is disclosed. Two dimensional images are projected to a three dimensional space then searched for matching attributes. Searching for corresponding attributes is much easier and computationally less intensive in the 3D space compared to a search in the 2D space. The method includes steps of projecting at least part of the images of said plurality to a 3-dimensional space resulting in a plurality of 3-D projected images (image_3D, image_3D), searching for at least one corresponding set of elements within the 3D projected images of the plurality of 3-D projected images, and calculating back said corresponding elements within the original images of the plurality and providing said corresponding elements within said original images as said at least one set of matched attributes.

Method and arrangement for improved image matching

A method for constructing an image model (M; M) from at least one image data input (IV; IV-IVn), comprises the steps of, in an iterative way,

Method and arrangement for image model construction

A method for determining calibration data for at least two cameras (camera, camera) in a multi view position, includes a step of determining respective parameters ((h, . . . , h), (h, . . . , h)) for identifying at least one respective homographic transformation on respective images (image,image) taken by said cameras of a same scene, by performing respective geometry analyses on said respective images (image, image), a step of performing at least one respective combined homographic transformation/feature detection step on said respective images thereby obtaining respective sets (feature set, feature set) of features on respective transformed images, such that said calibration data are obtained from matches (m, . . . , m) determined between said respective sets of features.

Method and arrangement for multi-camera calibration

Method for adapting a 3D model (m) of an object, said method comprising the steps of

Method and arrangement for 3-dimensional image model adaptation

An exemplary method of providing a virtual representation includes determining a plurality of position constraints that correspond to physical positions of individuals relative to each other. A plurality of determined viewing constraints each correspond to a position of one of the individuals relative to a display used by that individual for viewing the virtual representation. Relative positions of virtual representations of the individuals in the virtual representation are determined to correspond to the determined position and viewing constraints.

Method and system for providing consistency between a virtual representation and corresponding physical spaces

A feedback-system for managing the position of a frontal camera provided on a portable multimedia device during video communication. The feedback-system comprises a camera image analyzer coupled to the frontal camera and adapted to detect and analyze the pose of a user facing the frontal camera, an optimal video estimator coupled to the camera image analyzer and adapted to calculate a 6-dimensional error-vector of the current position of the frontal camera with respect to an optimal position of the frontal camera, an intuitive feedback manager coupled to the optimal video estimator and adapted to generate a transformation matrix translating the error-vector into an error message for the image displayed on a screen of the portable multimedia device, the intuitive feedback manager being further coupled to the portable multimedia device which is adapted to use the error message for modifying the image displayed on the screen.

Feedback-system for managing video conferencing with a portable multimedia device comprising a frontal camera

The invention pertains to a method for segmenting an image from a sequence of video images into a foreground and a background, the image being composed of pixels, the method includes assigning an initial foreground probability to each of the pixels; assigning a set of probability propagation coefficients to each of the pixels; and applying a global optimizer to the initial foreground probabilities of the pixels to classify each of the pixels as a foreground pixel or a background pixel, to obtain a definitive foreground map; wherein the global optimizer classifies each processed pixel in function of the initial foreground probability of the processed pixel and the initial foreground probability of neighboring pixels, the relative weight of the initial foreground probability of neighboring pixels being determined by the probability propagation coefficients of the neighboring pixels.

Method and system for segmenting an image

An exemplary technique is provided for obtaining an improved depth precision for a stereo image of an object of interest. The technique includes imaging a first image of the object of interest from a first position, imaging a second image of the object of interest from a second position different from the first position, determining a rough estimate of a depth for the object of interest based on the first image and the second image, determining an improved set of imaging positions corresponding to an improved depth precision for the object of interest based on the rough estimate of the depth, imaging improved images from the improved set of imaging positions, and determining an improved depth based on the improved images.

Method and systems for obtaining an improved stereo image of an object

A method for determining the relative position and orientation of a first and a second imaging device by comparing the image of a commonly depicted object in a first and a second image of the first and the second imaging device respectively, wherein the first and the second imaging devices are adapted for providing first and second depth information respectively for the respective images and in which at least the first and the second depth information is used for the determination. Corresponding device are also disclosed.

Growing community of inventors

Sint-Amandsberg, Belgium

Donny G Tytgat