摘要:
The invention relates to a device and a method for segmenting objects in an image. The method comprises steps for:
the approximate definition of a contour external to the object to be segmented, said contour including a plurality of so-called control points (V i ), the determination of a vicinity of candidate points (V i,j ) in the vicinity of each control point (V i ), the determination of an energy associated with each candidate point (V i,j ) in the vicinity of each control point, the convergence of each of the control points (V i ) towards the candidate point (Vi,j), which minimizes said energy, so providing an estimated contour of the object to be segmented.
According to the invention, the energy associated with each candidate point (Vi,j) includes a colour discrimination term.
摘要:
A method of displaying a neutral facial expression in a paused video includes analyzing contiguous frames of a video to find neutral facial expressions and associating one of the neutral facial expression frames with each frame in the video. Upon reception of a pause instruction, synthesizing a set of frames that transition from the associated neutral expression frame to a pause frame. The synthesized frames are played in reverse, whereby the played frames transition from the pause frame to the associated neutral expression frame. Thus, the neutral facial expression frame is displayed on the video player while the video is paused instead of displaying the normal pause frame.
摘要:
The invention relates to a guidance method for taking a picture, implemented by an apparatus comprising a display device (DD). According to the invention, the method comprises at least one iteration of : - obtaining (10) an image representative of a subject to be photographed, called a guidance image (GUI), via an acquisition device (DA); - comparing (20) of a set (ECGu) of at least one feature of said guidance image, called a guidance feature (CGu), with a corresponding set (ECSO) of at least one desired characteristic (CSO), delivering a comparison result (RESC);
and when said desired characteristic (CSO) differs from said guidance feature (CGu), displaying (30) on said display device (DD) of at least one item of correction information (Icorr) of said guidance image in view of said comparison result (RESC).
摘要:
The invention relates to a method of image processing adapted for tracking an object in a sequence of images. The method is based on two tracking methods (10, 34) that run independently and possibly exchange information (36, 38) at the end depending of the success of failure of said tracking methods. More precisely, according to the invention, the successful tracking method provides information in the form of a probability distribution of locations of the target object to the other tracking method.
摘要:
The application concerns a method for transmitting video data in a data stream and associated metadata, said metadata representing associated video data structure. The method comprises the steps of - determining the structure of the video data, - splitting said structure into chunks in an incremental manner, - transmitting said chunks in the data stream at time intervals before the video data it is related to.
摘要:
The invention relates to a device (5) for tracking an object in a sequence of frames. The tracking device comprises a set of local patch trackers which are interconnected by geometric interconnections for tracking local patches from one frame to the next. The tracking device further comprises control means (55) adapted to control the rigidity of the geometric interconnections. The invention also relates to a method for tracking an object in a sequence of frames.
摘要:
The present invention generally relates to a method for capturing the 3D motion of an object, an unmanned aerial vehicle and a motion capture system. The present invention discloses the use of a camera and a depth sensor onboard an Unmanned Aerial Vehicle (UAV) the trajectory of which is controlled in such a way that potentially occluded passive markers always remain in the field of view of the camera. In that way, the camera onboard the UAV provides complementary localization means for some of the markers placed on the object that may occasionally, because of occlusions, not be visible to a sufficient number of cameras of the optical motion capture system during a motion capture session, and as a result not localizable by the camera setup of the motion capture system.
摘要:
Method for colour correction of image sequences. A visual summary of the image sequences is generated (210), displaying at least one image from a plurality of the image sequences. Colour correction is applied (220) to a feature in one of the displayed images and images comprising at least one feature being close to the colour corrected feature according to a distance criterion are selected (230), displayed (240), and colour correction, based on the colour correction to the feature, is applied (250) to the features in the selected images. An apparatus is also provided.