Video predictive encoding device and system, video predictive decoding device and system
A video predictive encoding device includes an input module to receive pictures forming a video sequence, and an encoding module to encode the pictures by either intra prediction or inter prediction to generate compressed picture data, and to packetize the compressed image data along with packet header information. The packet header information includes a picture type. The encoding module determines the picture type so as to uniquely indicate whether encoded picture data is used for reference in decoding of another picture.
1. A video predictive encoding device comprising:
an input module executable to receive a plurality of pictures forming a video sequence, the plurality of pictures being separated into a plurality of temporal layers; and
an encoding module executable to encode a received picture to generate a compressed form of the picture, and encapsulate the compressed form of the picture in a network abstraction layer (NAL) unit with NAL unit header information,
wherein
the NAL unit header information includes a nal_unit_type that is variable to define different types of pictures and assign a first definition or a second definition, which is different from the first definition, to one picture of the video sequence so that with a selective assignment of the first or second definition to said one picture, the nal_unit_type is operable to (i) independently identify said one picture to be either a reference picture referenced in a prediction process by another picture of the video sequence residing in a temporal layer in which said one picture resides or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer and (ii) generate information to store only said reference picture in a memory, and
wherein the encoding module is executable to: determine whether said one picture is a reference picture referenced in a prediction process by another picture of the video sequence residing in the temporal layer or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer; in response to a determination that said one picture is a reference picture referenced in a prediction process by another picture of the video sequence residing in the temporal layer, define the nal_unit_type to assign said one picture the first definition; and in response to a determination that said one picture is a non-reference picture not referenced in a prediction process by any picture of the video sequence residing in the temporal layer, define the nal_unit_type to assign said one picture the second definition.
2. A video predictive decoding device comprising:
an input module executable to receive a compressed form of a plurality of pictures forming a video sequence, the plurality of pictures being separated into a plurality of temporal layers, wherein the compressed form of a picture is encapsulated in a network abstraction layer (NAL) unit with NAL unit header information; and
a decoding module executable to decode the NAL unit including the NAL unit header information and the compressed form of an encapsulated picture in the NAL unit,
wherein
the NAL unit header information includes a nal_unit_type that is variable to define different types of pictures and assign a first definition or a second definition, which is different from the first definition, to the encapsulated picture of the video sequence so that with a selective assignment of the first or second definition to the encapsulated picture, the nal_unit_type is operable to (i) independently identify the encapsulated picture to be either a reference picture referenced in a prediction process by another picture of the video sequence residing in a temporal layer in which the encapsulated picture resides or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer and (ii) generate information to store only said reference picture in a memory, and
wherein the decoding module is executable to: determine a definition defined in the nal_unit_type of the encapsulated picture; in response to a determination that the nal_unit_type assigns the encapsulated picture the first definition, allow the encapsulated picture to be used as a reference picture to decode another picture of the video sequence residing in the temporal layer and store only the encapsulated picture in the memory for decoding; and in response to a determination that the nal_unit_type assigns the encapsulated picture the second definition, disallow the encapsulated picture to be used as a reference picture to decode any picture of the video sequence residing in the temporal layer.
3. A video predictive encoding method comprising:
an input step of receiving a plurality of pictures forming a video sequence, the plurality of pictures being separated into a plurality of temporal layers; and
an encoding step of encoding a received picture to generate a compressed form of the picture, and encapsulating the compressed form of the picture in a network abstraction layer (NAL) unit with NAL unit header information,
wherein
the NAL unit header information includes a nal_unit_type that is variable to define different types of pictures and assign a first definition or a second definition, which is different from the first definition, to one picture of the video sequence so that with a selective assignment of the first or second definition to said one picture, the nal_unit_type is operable to (i) independently identify said one picture to be either a reference picture referenced in a prediction process by another picture of the video sequence residing in a temporal layer in which said one picture resides or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer and (ii) generate information to store only said reference picture in a memory, and
wherein the encoding step further comprises: determining whether said one picture is a reference picture referenced in a prediction process by another picture of the video sequence residing in the temporal layer or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer; in response to a determination that said one picture is a reference picture referenced in a prediction process by another picture of the video sequence residing in the temporal layer, defining the nal_unit_type to assign said one picture the first definition; and in response to a determination that said one picture is a non-reference picture not referenced in a prediction process by any picture of the video sequence residing in the temporal layer, defining the nal_unit_type to assign said one picture the second definition.
4. A video predictive decoding method comprising:
an input step of receiving a compressed form of a plurality of pictures forming a video sequence, the plurality of pictures being separated into a plurality of temporal layers, wherein the compressed form of a picture is encapsulated in a network abstraction layer (NAL) unit with NAL unit header information; and
a decoding step of decoding the NAL unit including the NAL unit header information and the compressed form of an encapsulated picture in the NAL unit,
wherein
the NAL unit header information includes a nal_unit_type that is variable to define different types of pictures and assign a first definition or a second definition, which is different from the first definition, to the encapsulated picture of the video sequence so that with a selective assignment of the first or second definition to the encapsulated picture, the nal_unit_type is operable to (i) independently identify the encapsulated picture to be either a reference picture referenced in a prediction process by another picture of the video sequence residing in a temporal layer in which the encapsulated picture resides or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer and (ii) generate information to store only said reference picture in a memory, and
wherein the decoding step further comprises: determining a definition defined in the nal_unit_type of the encapsulated picture; in response to a determination that the nal_unit_type assigns the encapsulated picture the first definition, allowing the encapsulated picture to be used as a reference picture to decode another picture of the video sequence residing in the temporal layer and storing only the encapsulated picture in the memory for decoding; and in response to a determination that the nal_unit_type assigns the encapsulated picture the second definition, disallowing the encapsulated picture to be used as a reference picture to decode any picture of the video sequence residing in the temporal layer.
5. A non-transitory computer readable storage medium comprising instructions executed by a computer for video predictive encoding, the computer readable storage medium comprising:
instructions executable to receive a plurality of pictures forming a video sequence, the plurality of pictures being separated into a plurality of temporal layers;
instructions executable to encode a received picture to generate a compressed form of the picture, and encapsulate the compressed form of the picture in a network abstraction layer (NAL) unit with NAL unit header information, wherein the NAL unit header information includes a nal_unit_type that is variable to define different types of pictures and assign a first definition or a second definition, which is different from the first definition, to one picture of the video sequence so that with a selective assignment of the first or second definition to said one picture, the nal_unit_type is operable to (i) independently identify said one picture to be either a reference picture referenced in a prediction process by another picture residing in a temporal layer in which said one picture resides or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer and (ii) generate information to store only said reference picture in a memory, and
instructions executable to encode the pictures further comprise instructions executable to: determine whether said one picture is a reference picture referenced in a prediction process by another picture of the video sequence residing in the temporal layer or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer; in response to a determination that said one picture is a reference picture referenced in a prediction process by another picture of the video sequence residing in the temporal layer, define the nal_unit_type to assign said one picture the first definition; and in response to a determination that said one picture is a non-reference picture not referenced in a prediction process by any picture of the video sequence residing in the temporal layer, define the nal_unit_type to assign said one picture the second definition.
6. A non-transitory computer readable storage medium comprising instructions executed by a computer for video predictive decoding, the computer readable storage medium comprising:
instructions executable to receive a compressed form of a plurality of pictures forming a video sequence, the plurality of pictures being separated into a plurality of temporal layers, wherein the compressed form of a picture is encapsulated in a network abstraction layer (NAL) unit with NAL unit header information; and
instructions executable to decode the NAL unit including the NAL unit header information and the compressed form of an encapsulated picture in the NAL unit,
wherein
the NAL unit header information includes a nal_unit_type that is variable to define different types of pictures and assign a first definition or a second definition, which is different from the first definition, to the encapsulated picture of the video sequence so that with a selective assignment of the first or second definition to the encapsulated picture, the nal_unit_type is operable to (i) independently identify the encapsulated picture to be either a reference picture referenced in a prediction process by another picture of the video sequence residing in a temporal layer in which the encapsulated picture resides or a non-reference picture not referenced by any picture of the video sequence residing in the temporal layer and (ii) generate information to store only said reference picture in a memory, and
wherein the instructions executable to: determine a definition defined in the nal_unit_type of the encapsulated picture, in response to a determination that the nal_unit_type assigns the encapsulated picture the first definition, allow the encapsulated picture to be used as a reference picture to decode another picture of the video sequence residing in the temporal layer and store only the encapsulated picture in the memory for decoding; and in response to a determination that the nal_unit_type assigns the encapsulated picture the second definition, disallow the encapsulated picture to be used as a reference picture to decode any picture of the video sequence residing in the temporal layer.