The invention relates to a method of selecting data of a compressed digital signal comprising a plurality of levels of spatial granularity of data, in a communication network comprising at least two communication apparatuses connected together by the network, the digital signal being available at least at one, so-called first, of the communication apparatuses, the data being adapted to be transmitted from the first apparatus to the other, so-called second, communication apparatus, and that data being sufficient to reconstitute a spatial part of the signal termed region of interest which is specified at the second apparatus, characterized in that the method comprises the following steps: determining (S76; S78; S80) at least one level of spatial granularity of data as a function of the region of interest and of the structure and organization of the data in the signal; selecting (S77; S79; S81) data for each determined level of granularity as a function of the region of interest and of the structure and organization of the data in the signal.