Method for encoding an input video comprising a luma component and two chroma components, the method comprising reshaping of said input video based on reshaping functions

A method comprising:

• decoding a reconstructed video from a bitstream, wherein the decoding comprises:
• obtaining a decoded reshaped video by decoding the bitstream;
• and
• obtaining the reconstructed video by inverse-reshaping the decoded reshaped video;
• wherein the inverse-reshaping of the decoded reshaped video depends on a targeted range of the decoded reshaped video and a range of a reconstructed video after the inverse-reshaping is applied on the decoded reshaped video, wherein the inverse-reshaping increases a range of codewords representing at least one component of the decoded reshaped video.

A device for encoding an input video comprising a luma component and two chroma components represented by code words, the device comprising a processor configured to:

• reshape the input video by applying a piece-wise linear function to the luma component and two other piece-wise linear functions to the chroma components to extend a range of the code words to a full range;
• encode the reshaped video into a bitstream;
• and
• encode parameters representative of the piece-wise linear functions as side information;
• wherein reshaping of the input video depends on a targeted range of a decoded reshaped video and a range of code words representing a reconstructed video after an inverse-reshaping is applied on the decoded reshaped video.