Abstract: Embodiments of this disclosure disclose an image data processing method and apparatus. The processing method includes: processing a first image and a second image by using a first neural network to obtain a homography matrix; determining a mapped image feature of a first image feature based on the homography matrix; fusing the mapped image feature and a second image feature to obtain a fused image feature; and processing the fused image feature by using a second neural network to obtain a first pixel height-to-depth ratio of the second image. In the embodiments of this disclosure, a dense and accurate pixel height-to-depth ratio can be obtained, thereby facilitating 3D scene modeling.

Abstract: In a point cloud attribute decoding method and device, Hilbert order is used as the coding order of point cloud attributes. The nearest neighbor points in the three-dimensional space of the current point are found in several points before the current point and are used as the prediction points of the current point. In the worst case, the distance in the three-dimensional space of the two points with a distance of n on the Hilbert order is less than or equal to n1/2, while the distance in the three-dimensional space of the two points with a distance of n on the Morton order may be greater than the maximum dimension length of the point cloud. The average distance between K points in the preceding order and the current point in Hilbert sequence is smaller than that between K points in the preceding order and the current point in Morton sequence.

Abstract: The present disclosure relates to compounds for RNA capping and uses thereof, which belong to the technical field of genetic engineering. Compounds of formula I can result in high levels of capping efficiency, and the capped mRNA exhibits a high transcription level and increased expression.

Abstract: A data processing method is performed by a receiving device. The method includes: receiving, through at least two network channels, data elements transmitted by a transmitting device, the data elements comprising delivery serial numbers; obtaining subsequent data elements of the received nth data element on the basis of the delivery serial numbers; and determining, when the subsequent data elements comprise a target data element, that a delivery result of the nth data element is discarded, a state of the target data element being a complete state, the target data element having a same data type as the data type of the nth data element, the target data element being an independent data element, the data type being determined on the basis of an effect of the data element, and a dependence degree of the independent data element on other data elements during post-processing being less than a specified degree.

Abstract: The present disclosure discloses a suction toy, including a shell, a printed circuit board (PCB), an actuating device, a piston, and a power supply. The shell is provided with a piston chamber; the piston chamber is communicated with an air inlet channel and an air outlet channel; the air inlet channel is provided with an air inlet, and the air outlet channel is provided with an air outlet; the shell is provided with a suction mouth; and the air inlet and the air outlet are both communicated with the suction mouth. Therefore, compared with the prior art, the suction toy can generate larger air flow and provide users with a better experience, and also has the advantages of difficulty in damage and longer service life.

Abstract: A data encoding method includes obtaining an attribute residual of a current point cloud point, binarizing the attribute residual to obtain a binary code of the current point cloud point that includes a first binary code indicating a first flag bit and a second binary code indicating a second flag bit, selecting a first context model from a context model list according to a first condition, selecting a second context model from the context model list according to a second condition, encoding the first binary code using the first context model, and encoding the second binary code using the second context model. The first condition and the second condition are different for the first context model and the second context model corresponding to a same index in the context model list.

Abstract: A data encoding method includes determining a value of a syntax element of point cloud data of a point cloud. The point cloud data includes attribute values of the point cloud and the syntax element indicates a search range of the attribute values during prediction encoding. The method further includes performing prediction processing on the attribute values according to the search range to obtain residuals of the attribute values, and encoding the residuals and a difference between the value of the syntax element and a constant value to generate code stream data. The constant value is a positive number.

Abstract: A point cloud attribute encoding method includes performing binarization on an attribute value in point cloud data to obtain a binary attribute value, and performing arithmetic encoding on bits in the binary attribute value using at least one probability model. A bit depth of the binary attribute value is larger than or equal to 1.