| CPC H04N 19/90 (2014.11) [H04N 19/172 (2014.11)] | 17 Claims |

|
1. A method for encoding feature data, comprising:
determining first feature data corresponding to each of channels to be encoded;
performing data type conversion processing on the first feature data corresponding to each channel to obtain second feature data that meets a data input condition of an apparatus for encoding feature data;
performing spatial re-expression on the second feature data corresponding to each channel to obtain third feature data, wherein a height of the third feature data conforms to a first height, a width of the third feature data conforms to a first width, and the third feature data comprises the second feature data located at a target position of each of the channels; and
encoding the third feature data and signalling in a bitstream,
wherein performing spatial re-expression on the second feature data corresponding to each channel to obtain the third feature data comprises:
padding the second feature data corresponding to each channel in spatial domain based on a second height and a second width to obtain padded second feature data;
splicing the padded second feature data corresponding to each channel into a target feature map in spatial domain, wherein the target feature map has the first height and the first width; and
acquiring feature data of the target feature map to obtain the third feature data.
|