[FFmpeg-devel] libopenjpeg decoder not correctly setting the pixel format for cinema JP2K wrapped MXF

Rémi Achard remiachard at gmail.com
Wed Sep 2 22:02:58 EEST 2020


Small update to the previous patch, adding a filter on
JPEG2000PictureSubDescriptor when iterating through the list of
SubDescriptors.

diff --git a/libavformat/mxf.h b/libavformat/mxf.h
index fc587f19f0..3fb3c6d74d 100644
--- a/libavformat/mxf.h
+++ b/libavformat/mxf.h
@@ -49,6 +49,7 @@ enum MXFMetadataSetType {
     TaggedValue,
     TapeDescriptor,
     AVCSubDescriptor,
+    JPEG2000PictureSubDescriptor,
 };

 enum MXFFrameLayout {
diff --git a/libavformat/mxfdec.c b/libavformat/mxfdec.c
index 6f6e8d586c..8eac7fc944 100644
--- a/libavformat/mxfdec.c
+++ b/libavformat/mxfdec.c
@@ -325,9 +325,11 @@ static const uint8_t mxf_encrypted_essence_container[]
    = { 0x06,0x0e,0x2b,0x
 static const uint8_t mxf_random_index_pack_key[]           = {
0x06,0x0e,0x2b,0x34,0x02,0x05,0x01,0x01,0x0d,0x01,0x02,0x01,0x01,0x11,0x01,0x00
};
 static const uint8_t mxf_sony_mpeg4_extradata[]            = {
0x06,0x0e,0x2b,0x34,0x04,0x01,0x01,0x01,0x0e,0x06,0x06,0x02,0x02,0x01,0x00,0x00
};
 static const uint8_t mxf_avid_project_name[]               = {
0xa5,0xfb,0x7b,0x25,0xf6,0x15,0x94,0xb9,0x62,0xfc,0x37,0x17,0x49,0x2d,0x42,0xbf
};
-static const uint8_t mxf_jp2k_rsiz[]                       = {
0x06,0x0e,0x2b,0x34,0x02,0x05,0x01,0x01,0x0d,0x01,0x02,0x01,0x01,0x02,0x01,0x00
};
+static const uint8_t mxf_jp2k_rsiz[]                       = {
0x06,0x0e,0x2b,0x34,0x01,0x01,0x01,0x0a,0x04,0x01,0x06,0x03,0x01,0x00,0x00,0x00
};
 static const uint8_t mxf_indirect_value_utf16le[]          = {
0x4c,0x00,0x02,0x10,0x01,0x00,0x00,0x00,0x00,0x06,0x0e,0x2b,0x34,0x01,0x04,0x01,0x01
};
 static const uint8_t mxf_indirect_value_utf16be[]          = {
0x42,0x01,0x10,0x02,0x00,0x00,0x00,0x00,0x00,0x06,0x0e,0x2b,0x34,0x01,0x04,0x01,0x01
};
+static const uint8_t mxf_subdescriptor_array_smpte[]       = {
0x06,0x0e,0x2b,0x34,0x01,0x01,0x01,0x09,0x06,0x01,0x01,0x04,0x06,0x10,0x00,0x00
};
+static const uint8_t mxf_subdescriptor_array_interop[]     = {
0x06,0x0e,0x2b,0x34,0x01,0x01,0x01,0x02,0x06,0x01,0x01,0x04,0x06,0x10,0x00,0x00
};

 #define IS_KLV_KEY(x, y) (!memcmp(x, y, sizeof(y)))

@@ -1272,6 +1274,11 @@ static int mxf_read_generic_descriptor(void *arg,
AVIOContext *pb, int tag, int
                 rsiz == FF_PROFILE_JPEG2000_DCINEMA_4K)
                 descriptor->pix_fmt = AV_PIX_FMT_XYZ12;
         }
+        if (IS_KLV_KEY(uid, mxf_subdescriptor_array_smpte)
+         || IS_KLV_KEY(uid, mxf_subdescriptor_array_interop)) {
+            mxf_read_strong_ref_array(pb,
&descriptor->sub_descriptors_refs,
+
 &descriptor->sub_descriptors_count);
+        }
         break;
     }
     return 0;
@@ -2498,6 +2505,16 @@ static int mxf_parse_structural_metadata(MXFContext
*mxf)
                 }
             }

+            if (st->codecpar->codec_id == AV_CODEC_ID_JPEG2000) {
+                MXFDescriptor *desc = NULL;
+                for (k = 0; k < descriptor->sub_descriptors_count; k++) {
+                    if ((desc = mxf_resolve_strong_ref(mxf,
&descriptor->sub_descriptors_refs[k], JPEG2000PictureSubDescriptor))) {
+                        st->codecpar->format = desc->pix_fmt;
+                        break;
+                    }
+                }
+            }
+
             if (st->codecpar->codec_id == AV_CODEC_ID_RAWVIDEO) {
                 st->codecpar->format = descriptor->pix_fmt;
                 if (st->codecpar->format == AV_PIX_FMT_NONE) {
@@ -2753,6 +2770,7 @@ static const MXFMetadataReadTableEntry
mxf_metadata_read_table[] = {
     { {
0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x5c,0x00
}, mxf_read_generic_descriptor, sizeof(MXFDescriptor), Descriptor }, /*
VANC/VBI - SMPTE 436M */
     { {
0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x5e,0x00
}, mxf_read_generic_descriptor, sizeof(MXFDescriptor), Descriptor }, /*
MPEG2AudioDescriptor */
     { {
0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x64,0x00
}, mxf_read_generic_descriptor, sizeof(MXFDescriptor), Descriptor }, /* DC
Timed Text Descriptor */
+    { {
0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x5a,0x00
}, mxf_read_generic_descriptor, sizeof(MXFDescriptor),
JPEG2000PictureSubDescriptor }, /* JPEG2000PictureSubDescriptor */
     { {
0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x3A,0x00
}, mxf_read_track, sizeof(MXFTrack), Track }, /* Static Track */
     { {
0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x3B,0x00
}, mxf_read_track, sizeof(MXFTrack), Track }, /* Generic Track */
     { {
0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x14,0x00
}, mxf_read_timecode_component, sizeof(MXFTimecodeComponent),
TimecodeComponent },

Le mar. 1 sept. 2020 à 23:43, Rémi Achard <remiachard at gmail.com> a écrit :

> > I'll continue to work on trying to fix mxfdec.c anyway, thanks.
>
> Update to the patch proposal, seems to be working ok (as far as pixel
> format is concerned) for both SMPTE & INTEROP DCP.
>
>
> diff --git a/libavformat/mxfdec.c b/libavformat/mxfdec.c
> index 6f6e8d586c..4c146a152d 100644
> --- a/libavformat/mxfdec.c
> +++ b/libavformat/mxfdec.c
> @@ -325,9 +325,11 @@ static const uint8_t
> mxf_encrypted_essence_container[]     = { 0x06,0x0e,0x2b,0x
>  static const uint8_t mxf_random_index_pack_key[]           = {
> 0x06,0x0e,0x2b,0x34,0x02,0x05,0x01,0x01,0x0d,0x01,0x02,0x01,0x01,0x11,0x01,0x00
> };
>  static const uint8_t mxf_sony_mpeg4_extradata[]            = {
> 0x06,0x0e,0x2b,0x34,0x04,0x01,0x01,0x01,0x0e,0x06,0x06,0x02,0x02,0x01,0x00,0x00
> };
>  static const uint8_t mxf_avid_project_name[]               = {
> 0xa5,0xfb,0x7b,0x25,0xf6,0x15,0x94,0xb9,0x62,0xfc,0x37,0x17,0x49,0x2d,0x42,0xbf
> };
> -static const uint8_t mxf_jp2k_rsiz[]                       = {
> 0x06,0x0e,0x2b,0x34,0x02,0x05,0x01,0x01,0x0d,0x01,0x02,0x01,0x01,0x02,0x01,0x00
> };
> +static const uint8_t mxf_jp2k_rsiz[]                       = {
> 0x06,0x0e,0x2b,0x34,0x01,0x01,0x01,0x0a,0x04,0x01,0x06,0x03,0x01,0x00,0x00,0x00
> };
>  static const uint8_t mxf_indirect_value_utf16le[]          = {
> 0x4c,0x00,0x02,0x10,0x01,0x00,0x00,0x00,0x00,0x06,0x0e,0x2b,0x34,0x01,0x04,0x01,0x01
> };
>  static const uint8_t mxf_indirect_value_utf16be[]          = {
> 0x42,0x01,0x10,0x02,0x00,0x00,0x00,0x00,0x00,0x06,0x0e,0x2b,0x34,0x01,0x04,0x01,0x01
> };
> +static const uint8_t mxf_subdescriptor_array_smpte[]       = {
> 0x06,0x0e,0x2b,0x34,0x01,0x01,0x01,0x09,0x06,0x01,0x01,0x04,0x06,0x10,0x00,0x00
> };
> +static const uint8_t mxf_subdescriptor_array_interop[]     = {
> 0x06,0x0e,0x2b,0x34,0x01,0x01,0x01,0x02,0x06,0x01,0x01,0x04,0x06,0x10,0x00,0x00
> };
>
>  #define IS_KLV_KEY(x, y) (!memcmp(x, y, sizeof(y)))
>
> @@ -1272,6 +1274,11 @@ static int mxf_read_generic_descriptor(void *arg,
> AVIOContext *pb, int tag, int
>                  rsiz == FF_PROFILE_JPEG2000_DCINEMA_4K)
>                  descriptor->pix_fmt = AV_PIX_FMT_XYZ12;
>          }
> +        if (IS_KLV_KEY(uid, mxf_subdescriptor_array_smpte)
> +         || IS_KLV_KEY(uid, mxf_subdescriptor_array_interop)) {
> +            mxf_read_strong_ref_array(pb,
> &descriptor->sub_descriptors_refs,
> +
>  &descriptor->sub_descriptors_count);
> +        }
>          break;
>      }
>      return 0;
> @@ -2498,6 +2505,16 @@ static int mxf_parse_structural_metadata(MXFContext
> *mxf)
>                  }
>              }
>
> +            if (st->codecpar->codec_id == AV_CODEC_ID_JPEG2000) {
> +                MXFDescriptor *desc = NULL;
> +                for (k = 0; k < descriptor->sub_descriptors_count; k++) {
> +                    if ((desc = mxf_resolve_strong_ref(mxf,
> &descriptor->sub_descriptors_refs[k], SubDescriptor))) {
> +                        st->codecpar->format = desc->pix_fmt;
> +                        break;
> +                    }
> +                }
> +            }
> +
>              if (st->codecpar->codec_id == AV_CODEC_ID_RAWVIDEO) {
>                  st->codecpar->format = descriptor->pix_fmt;
>                  if (st->codecpar->format == AV_PIX_FMT_NONE) {
> @@ -2753,6 +2770,7 @@ static const MXFMetadataReadTableEntry
> mxf_metadata_read_table[] = {
>      { {
> 0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x5c,0x00
> }, mxf_read_generic_descriptor, sizeof(MXFDescriptor), Descriptor }, /*
> VANC/VBI - SMPTE 436M */
>      { {
> 0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x5e,0x00
> }, mxf_read_generic_descriptor, sizeof(MXFDescriptor), Descriptor }, /*
> MPEG2AudioDescriptor */
>      { {
> 0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x64,0x00
> }, mxf_read_generic_descriptor, sizeof(MXFDescriptor), Descriptor }, /* DC
> Timed Text Descriptor */
> +    { {
> 0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x5a,0x00
> }, mxf_read_generic_descriptor, sizeof(MXFDescriptor), SubDescriptor }, /*
> JPEG2000PictureSubDescriptor */
>      { {
> 0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x3A,0x00
> }, mxf_read_track, sizeof(MXFTrack), Track }, /* Static Track */
>      { {
> 0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x3B,0x00
> }, mxf_read_track, sizeof(MXFTrack), Track }, /* Generic Track */
>      { {
> 0x06,0x0e,0x2b,0x34,0x02,0x53,0x01,0x01,0x0d,0x01,0x01,0x01,0x01,0x01,0x14,0x00
> }, mxf_read_timecode_component, sizeof(MXFTimecodeComponent),
> TimecodeComponent },
>
> Le mar. 1 sept. 2020 à 18:59, Rémi Achard <remiachard at gmail.com> a écrit :
>
>> > Sure, here are some comparisons I just made, please note the huge
>> difference in speed between current master and n4.0 that I picked as a
>> representative of the state of the decoder back then.
>>
>> Also worth noting that the n4.0 decoder was not even able to transcode
>> the full sequence properly and output a huge amount of warnings / errors.
>>
>>>


More information about the ffmpeg-devel mailing list