Random access with advanced decoded picture buffer (dpb) management in video coding

Abstract

As one example, techniques for decoding video data include receiving a bitstream that includes one or more pictures of a coded video sequence (CVS), decoding a first picture according to a decoding order, wherein the first picture is a random access point (RAP) picture that is not an instantaneous decoding refresh (IDR) picture, and decoding at least one other picture following the first picture according to the decoding order based on the decoded first picture. As another example, techniques for encoding video data include generating a bitstream that includes one or more pictures of a CVS, wherein a first picture according to the decoding order is a RAP picture that is not an IDR picture, and avoiding including at least one other picture, other than the first picture, that corresponds to a leading picture associated with the first picture, in the bitstream.

What is claimed is: 
     
         1 . A method of decoding video data, the method comprising:
 receiving a bitstream comprising one or more pictures of a coded video sequence (CVS);   decoding a first picture of the one or more pictures according to a decoding order associated with the CVS, wherein the first picture is a random access point (RAP) picture that is not an instantaneous decoding refresh (IDR) picture;   decoding at least one of the one or more pictures, other than the first picture, following the first picture according to the decoding order, based on the decoded first picture;   identifying at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS; and   decoding the at least one of the one or more pictures, wherein decoding each of the at least one of the one or more pictures includes:
 identifying one or more reference pictures used to encode the respective picture; 
 determining whether any of the identified one or more reference pictures is unavailable to be decoded; 
 for each of the identified one or more reference pictures that is determined to be unavailable to be decoded, generating a virtual reference picture; and 
 decoding the respective picture based on the corresponding one or more generated virtual reference pictures. 
   
     
     
         2 . The method of  claim 1 , wherein generating the virtual reference picture comprises generating the virtual reference picture using pixel values that correspond to a middle of a range of pixel values associated with the CVS. 
     
     
         3 . The method of  claim 2 , wherein the pixel values used to generate the virtual reference picture comprise luma intensity values that correspond to a middle of a range of luma intensity values associated with the CVS. 
     
     
         4 . The method of  claim 3 ,
 wherein generating the virtual reference picture using the luma values comprises generating the virtual reference picture using one or more instances of a luma intensity value of seventeen (17),   wherein a minimum luma intensity value of the range of luma intensity values associated with the CVS is zero (0), and   wherein a maximum luma intensity value of the range of luma intensity values associated with the CVS is two hundred fifty five (255).   
     
     
         5 . The method of  claim 2 , wherein the pixel values used to generate the virtual reference picture comprise chroma values that correspond to a middle of a range of chroma values associated with the CVS. 
     
     
         6 . The method of  claim 5 , wherein the virtual reference picture comprises a gray picture with respect to the CVS. 
     
     
         7 . The method of  claim 1 , further comprising:
 identifying at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS; and   avoiding outputting one or more of the at least one of the one or more pictures for which an output flag indicates that the respective picture is to be output.   
     
     
         8 . The method of  claim 1 , further comprising:
 identifying at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS; and   avoiding using one or more of the at least one of the one or more pictures as a reference picture for decoding at least one of the one or more pictures, other than the first picture, that follows the first picture according to the decoding order and according to a display order associated with the CVS.   
     
     
         9 . The method of  claim 1 , wherein the first picture comprises a clean random access (CRA) picture, wherein the CRA picture comprises a picture that is coded using intra-prediction coding and is decodable without reference to any other pictures, and for which one or more pictures included within a CVS along with the CRA picture that follow the CRA picture according to a decoding order associated with the CVS may be decoded with reference to one or more pictures that precede the CRA picture according to the decoding order. 
     
     
         10 . The method of  claim 1 , wherein the IDR picture comprises a picture that is coded using intra-prediction coding and is decodable without reference to any other pictures, and for which all other pictures included within a CVS along with the IDR picture that follow the IDR picture according to a decoding order associated with the CVS are decoded without reference to any pictures that precede the IDR picture according to the decoding order. 
     
     
         11 . A device for decoding encoded video data, the device comprising:
 a memory configured to store at least a portion of the encoded video data; and   a video decoder configured to:   receive an encoded video bitstream comprising one or more pictures of a coded video sequence (CVS);   decode a first picture of the one or more pictures according to a decoding order associated with the CVS, wherein the first picture is a random access point (RAP) picture that is not an instantaneous decoding refresh (IDR) picture;   decode at least one of the one or more pictures, other than the first picture, following the first picture according to the decoding order, based on the decoded first picture;   identify at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS; and   decode the at least one of the one or more pictures, wherein to decode each of the at least one of the one or more pictures, the video decoder is configured to:
 identify one or more reference pictures used to encode the respective picture; 
 determine whether any of the identified one or more reference pictures is unavailable to be decoded; 
 for each of the identified one or more reference pictures that is determined to be unavailable to be decoded, generate a virtual reference picture; and 
 decode the respective picture based on the corresponding one or more generated virtual reference pictures. 
   
     
     
         12 . The device of  claim 11 , wherein to generate the virtual reference picture, the video decoder is configured to generate the virtual reference picture using pixel values that correspond to a middle of a range of pixel values associated with the CVS. 
     
     
         13 . The device of  claim 12 , wherein the pixel values used to generate the virtual reference picture comprise luma intensity values that correspond to a middle of a range of luma intensity values associated with the CVS. 
     
     
         14 . The device of  claim 13 ,
 wherein to generate the virtual reference picture using the luma values, the video decoder is configured to generate the virtual reference picture using one or more instances of a luma intensity value of seventeen (17),   wherein a minimum luma intensity value of the range of luma intensity values associated with the CVS is zero (0), and   wherein a maximum luma intensity value of the range of luma intensity values associated with the CVS is two hundred fifty five (255).   
     
     
         15 . The device of  claim 12 , wherein the pixel values used to generate the virtual reference picture comprise chroma values that correspond to a middle of a range of chroma values associated with the CVS. 
     
     
         16 . The device of  claim 15 , wherein the virtual reference picture comprises a gray picture with respect to the CVS. 
     
     
         17 . The device of  claim 11 , wherein the device comprises at least one of:
 an integrated circuit;   a microprocessor;   a display configured to output the decoded at least one picture; or   a wireless communication device that includes the video decoder.   
     
     
         18 . A method of encoding video data, the method comprising:
 generating a bitstream comprising one or more pictures of a coded video sequence (CVS), wherein a first picture of the one or more pictures according to a decoding order associated with the CVS is a random access point (RAP) picture that is not an instantaneous decoding refresh (IDR) picture,
 wherein generating the bitstream comprises avoiding including at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS, 
 wherein the first picture is decodable, and wherein at least one of the one or more pictures, other than the first picture, following the first picture according to the decoding order, is decodable based on the first picture, and 
 wherein encoding further includes a decoding process that comprises: 
 identifying at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS; and 
 decoding the at least one of the one or more pictures, wherein decoding each of the at least one of the one or more pictures includes:
 identifying one or more reference pictures used to encode the respective picture; 
 determining whether any of the identified one or more reference pictures is unavailable to be decoded; 
 for each of the identified one or more reference pictures that is determined to be unavailable to be decoded, generating a virtual reference picture; and 
 decoding the respective picture based on the corresponding one or more generated virtual reference pictures. 
 
   
     
     
         19 . A device for encoding video data, the device comprising:
 a memory configured to store at least a portion of the video data; and   a video encoder configured to:
 generate a bitstream comprising one or more pictures of a coded video sequence (CVS), wherein a first picture of the one or more pictures according to a decoding order associated with the CVS is a random access point (RAP) picture that is not an instantaneous decoding refresh (IDR) picture, 
 wherein to generate the bitstream, the video encoder is configured to avoid including at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS, 
 wherein the first picture is decodable, and wherein at least one of the one or more pictures, other than the first picture, following the first picture according to the decoding order, is decodable based on the first picture; and 
   a video decoder configured to:
 identify at least one of the one or more pictures, other than the first picture, that corresponds to a leading picture associated with the first picture, wherein the leading picture comprises a picture that follows the first picture according to the decoding order and precedes the first picture according to a display order associated with the CVS; and 
 decode the at least one of the one or more pictures, wherein to decode each of the at least one of the one or more pictures, the video decoder is configured to:
 identify one or more reference pictures used to encode the respective picture; 
 determine whether any of the identified one or more reference pictures is unavailable to be decoded; 
 for each of the identified one or more reference pictures that is determined to be unavailable to be decoded, generate a virtual reference picture; and 
 decode the respective picture based on the corresponding one or more generated virtual reference pictures. 
 
   
     
     
         20 . The device of  claim 19 , wherein the device comprises at least one of:
 an integrated circuit;   a microprocessor;   a camera configured to capture at least a portion of the video data; or   a wireless communication device that includes the video decoder.