Abstract: An input stereoscopic video signal has a first frame sequence (SL) in a left-eye channel and a second frame sequence (SR) in a right-eye channel, the second frame sequence (SR) having a zero or non-zero time offset ?1 with respect to the first frame sequence (SL). It is converted into an output stereoscopic video signal having a third frame sequence (S?L) in the left-eye channel and a fourth frame sequence (S?R) in the right-eye channel. The fourth frame sequence (S?R) has a zero or non-zero time offset ?2 with respect to the third left-eye frame sequence (S?L). The two time offsets ?1, ?2 may be different ?2??1), in which case the conversion comprises a time interpolation of the frames of at least one of the first and second frame sequences with interpolation parameters selected to apply a relative time shift of ?2??1 to the right-eye channel with respect to the left-eye channel.

Abstract: A method is proposed for analyzing an interlaced video signal including a first sequence of fields. A temporal regularity estimation process is applied to the first sequence of fields to compute a first metric. Inputs of the temporal regularity estimation process include pixel values from at least two fields having respective ranks differing by more than one in the sequence. The same temporal regularity estimation process is applied to second and third sequences of fields to compute second and third metrics. The second sequence is derived from the first sequence by swapping fields having ranks of the form 2k and 2k+1 for any integer k, while the third sequence is derived from the first sequence by swapping fields having ranks of the form 2k?1 and 2k. The first, second and third metrics are compared in a determination of the time arrangement of the fields in the first sequence.

Abstract: For compressing a video signal, a local multiscale transform is applied to a frame of the video signal to obtain coefficient blocks. The coefficients of each block are distributed into a plurality of coefficient groups, and for at least one of the groups, a common exponent is determined for encoding the coefficients of the group, and respective mantissas are determined for quantizing the coefficients of the group in combination with the common exponent. Coding data including each exponent determined for a coefficient group and the mantissas quantizing the coefficients of the group in combination with this exponent are stored in an external frame buffer.

Abstract: The method comprises for a pixel of an output grid which is not on an input grid on which pixel values are received in an input signal: determining a loss value for each of a plurality of candidate interpolators by analyzing the pixel values of the input grid along a direction associated with the candidate interpolator; selecting at least one interpolator by minimizing the loss value; and determining an output pixel value. At least two interpolation modes are provided, including: a first mode in which one interpolator is selected and the output pixel value is determined as an interpolated value obtained by applying locally the selected interpolator to pixel values of the input grid; and a second mode in which more than one interpolator is selected and the output pixel value is determined as a weighted sum of a plurality of interpolated values obtained by applying locally the selected interpolators.

Abstract: For compressing a video signal, a local multiscale transform is applied to a frame of the video signal to obtain coefficient blocks. The coefficients of each block are distributed into coefficient groups associated with that block. A plurality of the coefficient groups associated with a block are processed. The processing of one of the groups comprises determining an exponent for encoding the coefficients of that group. Mantissas are determined for quantizing the coefficients of the plurality of groups in combination with the exponents respectively determined for these groups. Coding data including each exponent determined for a coefficient group and the mantissas quantizing the coefficients of the group in combination with this exponent are stored in an external frame buffer. The mantissas determined for quantizing the coefficients of one of the groups are represented in the coding data by a respective number of bits depending on the exponents determined for the plurality of coefficient groups.

Abstract: For scaling an input image into an output image, the method comprises, for a point of the output sampling grid which is not in the input sampling grid: calculating a plurality of interpolated pixel values by applying respective interpolators; determining respective loss values associated with the interpolated pixel values; and providing a pixel value of the output image using at least one of the interpolated pixel values selected by minimizing the loss value.

Abstract: A direction of regularity, which minimizes a directional energy computed from pixel values of consecutive first and second frames of an input video sequence, is respectively associated with each pixel of the first frame and with each pixel of the second frame. Another direction of regularity (vz), which minimizes a directional energy computed from pixel values of the first and second frames, is also associated with an output pixel (z) of a frame of an output video sequence, located in time between the first and second frames. For processing such output pixel, the respective minimized directional energies for the output pixel, at least one pixel (z?) of the first frame and at least one pixel (z?) of the second frame are compared to control an interpolation performed to determine a value of the output pixel. The interpolation uses pixel values from at least one of the first and second frames of the input video sequence depending on the comparison of the minimized directional energies.

Abstract: Evolution of a scene represented in a video sequence of input frames is analyzed. Output pixels of an output frame having a time position intermediate between time positions of the input frames are computed by combining respective input pixels of the input frames.

Abstract: A wavelet transform is applied to successive images of a video sequence to obtain wavelet coefficients for each image, and directions of regularity are estimated in association with the wavelet coefficients. Recursive weights are also determined to be associated with the wavelet coefficients. An average multiscale bandlet coefficient associated with a wavelet coefficient for a current image is computed from this wavelet coefficient, the recursive weight associated therewith and a corresponding average multiscale bandlet coefficient computed for a previous image and associated with a wavelet coefficient offset according to the direction of regularity associated with the wavelet coefficient. The average multiscale bandlet coefficients can then be processed to produce an enhanced bandlet image to which an inverse wavelet transform is applied.

Abstract: When analyzing an image signal having pixel values defined on a sampling grid, the method detects at least one direction of regularity of the image signal in relation to a pixel of the sampling grid. This detection comprises computing a respective loss value associated with at least one direction in a set of directions, and selecting at least one direction of regularity by minimizing the loss value. The loss value associated with a direction (u, v), where u is a horizontal coordinate and v a vertical coordinate, has an axial loss component measuring variations of the pixel values in at least one linear array of pixels. This linear array is a horizontal array if |u/v|>1 and a vertical array if |u/v|<1.

Abstract: For compressing a video signal, a local multiscale transform is applied to a frame of the video signal to obtain coefficient blocks. The coefficients of each block are distributed into a plurality of coefficient groups, and for at least one of the groups, a common exponent is determined for encoding the coefficients of the group, and respective mantissas are determined for quantizing the coefficients of the group in combination with the common exponent. Coding data including each exponent determined for a coefficient group and the mantissas quantizing the coefficients of the group in combination with this exponent are stored in an external frame buffer.

Abstract: In a method of analyzing an input video sequence, pixels of synthesized images of an output video sequence are associated with respective directions of regularity belonging to a predefined set of directions. A first subset of candidate directions is determined from the predefined set of directions for a region of a first image of the output sequence. For a corresponding region of a second synthesized image of the output sequence following the first image, a second subset of candidate directions is determined from the predefined set of directions, based on images of the input sequence and the first subset of candidate directions. The directions of regularity for pixels of this region of the second synthesized image are detected from the second subset of candidate directions. The recursive determination of the subsets of candidate directions provides a sparse geometry for efficiently analyzing the video sequence.

Abstract: For subband decomposition of a d-dimensional input signal (S) into a number K of subband components (F1-F4), a filter bank has a filtering module (801) transforming the input signal (S) into 2d components including a low-frequency component (L) and 2d?1 higher-frequency components (F1), The 2d?1 higher-frequency components are oversampled, typically by a factor 2, compared to the low-frequency component. The low-frequency component can be further decomposed by means of another filtering module having a similar structure, and the process can be iterated over any number of scales. The reconstruction filter bank has a symmetric structure, with filtering modules adapted to the oversampling of the higher-frequency components. Such filter banks are well suited to various enhancement processing applied to the subband components such as thresholding, reduction of compression distortion, reduction of measurement noise, sharpness enhancement.

Abstract: Input pixel values associated with input pixel positions having a first spacing along a direction are received and processed to determine output pixel values associated with output pixel positions having a second spacing along the direction, from respective combinations of input pixel values weighted by coefficients given by an interpolation function. In a downscaling operation, the second spacing is larger than the first spacing, and the weighting coefficients represent values of the interpolation function expanded spatially by an expansion factor equal to the ratio of the second spacing to the first spacing.

Abstract: In a method of analyzing an input video sequence, pixels of synthesized images of an output video sequence are associated with respective directions of regularity belonging to a predefined set of directions. A first subset of candidate directions is determined from the predefined set of directions for a region of a first image of the output sequence. For a corresponding region of a second synthesized image of the output sequence following the first image, a second subset of candidate directions is determined from the predefined set of directions, based on images of the input sequence and the first subset of candidate directions. The directions of regularity for pixels of this region of the second synthesized image are detected from the second subset of candidate directions. The recursive determination of the subsets of candidate directions provides a sparse geometry for efficiently analyzing the video sequence.

Abstract: For compressing a video signal, a local multiscale transform is applied to a frame of the video signal to obtain coefficient blocks. The coefficients of each block are distributed into a plurality of coefficient groups, and for at least one of the groups, a common exponent is determined for encoding the coefficients of the group, and respective mantissas are determined for quantizing the coefficients of the group in combination with the common exponent. Coding data including each exponent determined for a coefficient group and the mantissas quantizing the coefficients of the group in combination with this exponent are stored in an external frame buffer.

Abstract: For estimating an illumination pattern generated by the plurality of light sources, a combination of contributions of the light sources is computed. The contribution of a light source comprises a combination of at least a first component, comprising a part of a reference profile aligned on a position of said light source, said part of the aligned reference profile extending within the illumination area, and a second component depending on a distance between the position of said light source and an edge of the illumination area.

Abstract: The method comprises for a pixel of an output grid which is not on an input grid on which pixel values are received in an input signal: determining a loss value for each of a plurality of candidate interpolators by analyzing the pixel values of the input grid along a direction associated with the candidate interpolator; selecting at least one interpolator by minimizing the loss value; and determining an output pixel value. At least two interpolation modes are provided, including: a first mode in which one interpolator is selected and the output pixel value is deter mined as an interpolated value obtained by applying locally the selected interpolator to pixel values of the input grid; and a second mode in which more than one interpolator is selected and the output pixel value is determined as a weighted sum of a plurality of interpolated values obtained by applying locally the selected interpolators.

Abstract: When analyzing an image signal having pixel values defined on a sampling grid, the method detects at least one direction of regularity of the image signal in relation to a pixel of the sampling grid. This detection comprises computing a respective loss value associated with at least one direction in a set of directions, and selecting at least one direction of regularity by minimizing the loss value. The loss value associated with a direction (u, v), where u is a horizontal coordinate and v a vertical coordinate, has an axial loss component measuring variations of the pixel values in at least one linear array of pixels. This linear array is a horizontal array |u/v|>1 and a vertical array |u/v|<1.

Abstract: For scaling an input image into an output image, the method comprises, for a point of the output sampling grid which is not in the input sampling grid: calculating a plurality of interpolated pixel values by applying respective interpolators; determining respective loss values associated with the interpolated pixel values; and providing a pixel value of the output image using at least one of the interpolated pixel values selected by minimizing the loss value.