The present disclosure provides a deep neural network (DNN)-based reconstruction method and apparatus for compressive video sensing (CVS). The method divides a video signal into a key frame and a non-key frame. The key frame is reconstructed by using an existing image reconstruction method. The non-key frame is reconstructed by using a special DNN according to the present disclosure. The neural network includes an adaptive sampling module, a multi-hypothesis prediction module, and a residual reconstruction module. The neural network makes full use of a spatio-temporal correlation of the video signal to sample and reconstruct the video signal. This ensures low time complexity of an algorithm while improving reconstruction quality. Therefore, the method in the present disclosure is applicable to a video sensing system with limited resources on a sampling side and high requirements for reconstruction quality and real-time performance.

The disclosure is directed to a method of color palette coding and an electronic device using the same method. The proposed method would include not limited to encoding, by using a processor, a bit stream to represent a color of a coding unit and storing the bit stream in a storage medium or transmitting the bit stream. The bit stream would include not limited to a binary bit representing a run flag, a N binary bit sequence representing up to 2.sup.N major colors in a coding unit with each of the possible values of the N binary bit sequence representing an unique major color index of N major color indices, and a M binary bit sequence representing up to 2.sup.M extended colors with each of the possible values of the M binary bit sequence representing an unique extended color index of M extended color indices.

The disclosure is directed to a method of color palette coding and an electronic device using the same method. The proposed method would include not limited to encoding, by using a processor, a bit stream to represent a color of a coding unit and storing the bit stream in a storage medium or transmitting the bit stream. The bit stream would include not limited to a binary bit representing a run flag, a N binary bit sequence representing up to 2.sup.N major colors in a coding unit with each of the possible values of the N binary bit sequence representing an unique major color index of N major color indices, and a M binary bit sequence representing up to 2.sup.M extended colors with each of the possible values of the M binary bit sequence representing an unique extended color index of M extended color indices.

Methods and apparatus provide cloud-based video encoding that generates encoded video data by one or more encoders in a cloud platform for a plurality of cloud encoding sessions. The methods and apparatus generate operational improvement tradeoff data in response to operational encoding metrics associated with the one or more encoders and change operational characteristics of the one or more encoders for at least one of the cloud encoding sessions based on the operational improvement tradeoff data.

Methods and apparatus provide cloud-based video encoding that generates encoded video data by one or more encoders in a cloud platform for a plurality of cloud encoding sessions. The methods and apparatus generate operational improvement tradeoff data in response to operational encoding metrics associated with the one or more encoders and change operational characteristics of the one or more encoders for at least one of the cloud encoding sessions based on the operational improvement tradeoff data.

A computer-implemented method includes receiving a spatial feature descriptor representing a binary string of comparison results between specified pairings of N patches in a region of an image, N being an integer greater than one. The method further includes compressing the spatial feature descriptor by determining a minimum number of bits needed to represent N, and generating a sorted array representing a ranking of the N patches based on the binary string of comparison results. Each element of the sorted array is associated with a corresponding patch of the N patches and stores a value representing a ranking of the patch relative to the other N−1 patches and coded using the minimum number of bits needed to represent N. The sorted array may be further compressed using a progressive bit representation decimation process.

Aspects of the disclosure provide methods and apparatuses for point cloud compression. In some examples, an apparatus for point cloud compression includes processing circuitry. The processing circuitry determines a plurality of original points in a point cloud that is quantized to a reconstructed position. The processing circuitry determines an attribute value of the reconstructed position based on attribute information of the plurality of original points. Further, the processing circuitry encodes the point cloud based on the determined attribute value of the reconstructed position.

In an example, a method of coding video data includes determining a first palette having first entries indicating first pixel values, determining, based on the first entries of the first palette, one or more second entries indicating second pixel values of a second palette, and coding pixels of a block of video data using the second palette.

In an example, a method of coding video data includes determining a first palette having first entries indicating first pixel values, determining, based on the first entries of the first palette, one or more second entries indicating second pixel values of a second palette, and coding pixels of a block of video data using the second palette.

An image coding method for coding an image on a block-by-block basis, includes: selecting, for each of a plurality of sub-blocks included in a coding-target block and each including a plurality of coefficients, a context for performing arithmetic coding on a parameter indicating a coding-target coefficient included in the sub-block from a context set corresponding to the sub-block, based on at least one reference coefficient located around the coding-target coefficient, the coding-target block being a transform unit; and performing arithmetic coding on the parameter indicating the coding-target coefficient using probability information about the selected context, wherein, in the selecting, the context is selected from the context set, the context set corresponding to a sum of (i) a value indicating a position in a horizontal direction of the sub-block in the coding-target block and (ii) a value indicating a position in a vertical direction of the sub-block in the coding-target block.