Method and apparatus for deblocking an image

A method for encoding an image comprising:

• reconstructing an image part;
• determining at least one boundary between a first block of samples and a second block of samples, the first block and the second block belonging to the part of an image;
• determining that the first block has a geometric partitioning mode performing a weighted combination of a first predictor obtained by a first inter uni-directional prediction and a second predictor obtained by a second inter uni-directional prediction;
• storing one of an inter bi-directional prediction mode or an inter uni-directional prediction mode as a prediction mode of the first block according to a weight of a sample in the first predictor, wherein the prediction mode of the first block is an inter uni-directional prediction mode in case the weight of the sample in the first predictor is equal to zero or a maximum weight;
• determining a boundary strength of a deblocking filter according to at least one of a prediction mode of the first block and a prediction mode of the second block;
• and
• filtering samples neighboring the at least one boundary between the first block and the second block, according to the boundary strength of the deblocking filter.

An apparatus for decoding an image, the apparatus comprising one or more processor configured to perform:

• decoding a part of an image part;
• determining at least one boundary between a first block of samples and a second block of samples, the first block and the second block belonging to the part of an image;
• determining that the first block has a geometric partitioning mode performing a weighted combination of a first predictor obtained by a first inter uni-directional prediction and a second predictor obtained by a second inter uni-directional prediction;
• storing one of an inter bi-directional prediction mode or an inter uni-directional prediction mode as a prediction mode of the first block according to a weight of a sample in the first predictor, wherein the prediction mode of the first block is an inter uni-directional prediction mode in case the weight of the sample in the first predictor is equal to zero or a maximum weight;
• determining a boundary strength of a deblocking filter according to at least one of a prediction mode of the first block and a prediction mode of the second block;
• and
• filtering samples neighboring the at least one boundary between the first block and the second block, according to the boundary strength of the deblocking filter.

An apparatus for encoding an image, the apparatus comprising one or more processor configured to perform:

• reconstructing an image part;
• determining at least one boundary between a first block of samples and a second block of samples, the first block and the second block belonging to the part of an image;
• determining that the first block has a geometric partitioning mode performing a weighted combination of a first predictor obtained by a first inter uni-directional prediction and a second predictor obtained by a second inter uni-directional prediction;
• storing one of an inter bi-directional prediction mode or an inter uni-directional prediction mode as a prediction mode of the first block according to a weight of a sample in the first predictor, wherein the prediction mode of the first block is an inter uni-directional prediction mode in case the weight of the sample in the first predictor is equal to zero or a maximum weight;
• determining a boundary strength of a deblocking filter according to at least one of a prediction mode of the first block and a prediction mode of the second block;
• and
• filtering samples neighboring the at least one boundary between the first block and the second block, according to the boundary strength of the deblocking filter.