Problem with -vf fillborders on 16 bit image

classic Classic list List threaded Threaded
22 messages Options
12
Reply | Threaded
Open this post in threaded view
|

Problem with -vf fillborders on 16 bit image

Ulf Zibis
Hi,

I do some experiences with 16 bit input and I discover a colour problem
(see attached screenshot).
The input image is from https://samples.ffmpeg.org/image-samples/

Here is what I did:
$ ./ffmpeg -i debug/flower-separated-contig-16.tif -vf
fillborders=6:6:6:6:smear debug/flower-contig-16_smear-6-6-6-6.tif
ffmpeg version N-93301-gf95aee2b72 Copyright (c) 2000-2019 the FFmpeg
developers
  built with gcc 7 (Ubuntu 7.3.0-27ubuntu1~18.04)
  configuration:
  libavutil      56. 26.100 / 56. 26.100
  libavcodec     58. 47.103 / 58. 47.103
  libavformat    58. 26.101 / 58. 26.101
  libavdevice    58.  6.101 / 58.  6.101
  libavfilter     7. 48.100 /  7. 48.100
  libswscale      5.  4.100 /  5.  4.100
  libswresample   3.  4.100 /  3.  4.100
Input #0, tiff_pipe, from 'debug/flower-separated-contig-16.tif':
  Duration: N/A, bitrate: N/A
    Stream #0:0: Video: tiff, rgba64be, 73x43 [SAR 1:1 DAR 73:43], 25
tbr, 25 tbn, 25 tbc
File 'debug/flower-contig-16_smear-6-6-6-6.tif' already exists.
Overwrite ? [y/N] y
Stream mapping:
  Stream #0:0 -> #0:0 (tiff (native) -> tiff (native))
Press [q] to stop, [?] for help
Output #0, image2, to 'debug/flower-contig-16_smear-6-6-6-6.tif':
  Metadata:
    encoder         : Lavf58.26.101
    Stream #0:0: Video: tiff, rgba64le, 73x43 [SAR 1:1 DAR 73:43],
q=2-31, 200 kb/s, 25 fps, 25 tbn, 25 tbc
    Metadata:
      encoder         : Lavc58.47.103 tiff
frame=    1 fps=0.0 q=-0.0 Lsize=N/A time=00:00:00.04 bitrate=N/A
speed=0.644x   
video:25kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB
muxing overhead: unknown

-Ulf

--
Von meinem Seibert gesendet


_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".

flower-separated-contig-16.tif (34K) Download Attachment
flower-contig-16_smear-6-6-6-6.tif (34K) Download Attachment
in+out_16bit.jpg (112K) Download Attachment
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Paul B Mahol
On 3/8/19, Ulf Zibis <[hidden email]> wrote:

> Hi,
>
> I do some experiences with 16 bit input and I discover a colour problem
> (see attached screenshot).
> The input image is from https://samples.ffmpeg.org/image-samples/
>
> Here is what I did:
> $ ./ffmpeg -i debug/flower-separated-contig-16.tif -vf
> fillborders=6:6:6:6:smear debug/flower-contig-16_smear-6-6-6-6.tif
> ffmpeg version N-93301-gf95aee2b72 Copyright (c) 2000-2019 the FFmpeg
> developers
>   built with gcc 7 (Ubuntu 7.3.0-27ubuntu1~18.04)
>   configuration:
>   libavutil      56. 26.100 / 56. 26.100
>   libavcodec     58. 47.103 / 58. 47.103
>   libavformat    58. 26.101 / 58. 26.101
>   libavdevice    58.  6.101 / 58.  6.101
>   libavfilter     7. 48.100 /  7. 48.100
>   libswscale      5.  4.100 /  5.  4.100
>   libswresample   3.  4.100 /  3.  4.100
> Input #0, tiff_pipe, from 'debug/flower-separated-contig-16.tif':
>   Duration: N/A, bitrate: N/A
>     Stream #0:0: Video: tiff, rgba64be, 73x43 [SAR 1:1 DAR 73:43], 25
> tbr, 25 tbn, 25 tbc
> File 'debug/flower-contig-16_smear-6-6-6-6.tif' already exists.
> Overwrite ? [y/N] y
> Stream mapping:
>   Stream #0:0 -> #0:0 (tiff (native) -> tiff (native))
> Press [q] to stop, [?] for help
> Output #0, image2, to 'debug/flower-contig-16_smear-6-6-6-6.tif':
>   Metadata:
>     encoder         : Lavf58.26.101
>     Stream #0:0: Video: tiff, rgba64le, 73x43 [SAR 1:1 DAR 73:43],
> q=2-31, 200 kb/s, 25 fps, 25 tbn, 25 tbc
>     Metadata:
>       encoder         : Lavc58.47.103 tiff
> frame=    1 fps=0.0 q=-0.0 Lsize=N/A time=00:00:00.04 bitrate=N/A
> speed=0.644x
> video:25kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB
> muxing overhead: unknown
>
> -Ulf
>
> --
> Von meinem Seibert gesendet
>
>

tiff is encoded with alpha, which is not initially available. Filter
is working fine.
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis

Am 09.03.19 um 00:18 schrieb Paul B Mahol:
> On 3/8/19, Ulf Zibis <[hidden email]> wrote:
>
>
> tiff is encoded with alpha, which is not initially available. Filter
> is working fine.
I do not really understand this argumentation, but this is not my main
interest.

It like to try some tweaks with fillborders filter, so I need some
16-bit per plane images for testing. Can you help me with that?

Another question is, why the fillborders doesn't check if the input is
writable, as suggested in doc/writing_filters.txt.

-Ulf

_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Michael Koch

> It like to try some tweaks with fillborders filter, so I need some
> 16-bit per plane images for testing. Can you help me with that?

You can make 16-bit images with the latest version of GIMP. Use "export
as", then use *.png as filename, then select "16bpc RGB" as pixel
format, then click on "export".

Michael
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Paul B Mahol
In reply to this post by Ulf Zibis
On 3/9/19, Ulf Zibis <[hidden email]> wrote:

>
> Am 09.03.19 um 00:18 schrieb Paul B Mahol:
>> On 3/8/19, Ulf Zibis <[hidden email]> wrote:
>>
>>
>> tiff is encoded with alpha, which is not initially available. Filter
>> is working fine.
> I do not really understand this argumentation, but this is not my main
> interest.
>
> It like to try some tweaks with fillborders filter, so I need some
> 16-bit per plane images for testing. Can you help me with that?
>
> Another question is, why the fillborders doesn't check if the input is
> writable, as suggested in doc/writing_filters.txt.
>

Because it request writtable frames.

> -Ulf
>
> _______________________________________________
> ffmpeg-user mailing list
> [hidden email]
> https://ffmpeg.org/mailman/listinfo/ffmpeg-user
>
> To unsubscribe, visit link above, or email
> [hidden email] with subject "unsubscribe".
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Carl Eugen Hoyos-2
In reply to this post by Michael Koch
2019-03-09 9:26 GMT+01:00, Michael Koch <[hidden email]>:
>
>> It like to try some tweaks with fillborders filter, so I need some
>> 16-bit per plane images for testing. Can you help me with that?
>
> You can make 16-bit images with the latest version of GIMP.

You can make such files with FFmpeg and links to two such
files were provided in this thread.

Does gimp allow to create 16bit yuv files as requested?

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Michael Koch

> Does gimp allow to create 16bit yuv files as requested?

No, it can create 16-bit PNG files with pixel formats RGB, GRAY, RGBA
and GRAYA.
Doesn't FFmpeg automatically convert 16-bit RGB to 16-bit YUV, if this
is required?

Michael

--
**********************************************
   ASTRO ELECTRONIC   Dipl.-Ing. Michael Koch
        Raabestr. 43   37412 Herzberg
           www.astro-electronic.de
   Tel. +49 5521 854265   Fax +49 5521 854266
**********************************************

_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis
In reply to this post by Carl Eugen Hoyos-2

Am 09.03.19 um 10:10 schrieb Carl Eugen Hoyos:
>
> You can make such files with FFmpeg
Which format should I use with option -f ?

> and links to two such files were provided in this thread.
The files from ticket #503 are only gray, and the JPEG's from ticket
#855 seem corrupted or at least not common standard, as Ubuntu picture
viewer isn't able to display them.

Anyway thanks for the links.

-Ulf


_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis
In reply to this post by Michael Koch

Am 09.03.19 um 10:23 schrieb Michael Koch:
>
> Doesn't FFmpeg automatically convert 16-bit RGB to 16-bit YUV, if this
> is required?

-vf fillborders doesn't require YUV, so no automatic conversion happens.

-Ulf

_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Carl Eugen Hoyos-2
In reply to this post by Ulf Zibis
2019-03-09 10:44 GMT+01:00, Ulf Zibis <[hidden email]>:
>
> Am 09.03.19 um 10:10 schrieb Carl Eugen Hoyos:
>>
>> You can make such files with FFmpeg
> Which format should I use with option -f ?

You should never need the option -f (except when reading
raw g.729 and similar rare raw formats), what did you try?

>> and links to two such files were provided in this thread.
> The files from ticket #503 are only gray,

Yes.

> and the JPEG's from ticket #855 seem corrupted

They are not corrupted as you can easily test with FFmpeg.

(The way I understood your question you asked for 16bit
images that can be read with FFmpeg, not for 16 bit
images that can be read with other software which is
much more difficult.)

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Carl Eugen Hoyos-2
In reply to this post by Ulf Zibis
2019-03-09 10:50 GMT+01:00, Ulf Zibis <[hidden email]>:
>
> Am 09.03.19 um 10:23 schrieb Michael Koch:
>>
>> Doesn't FFmpeg automatically convert 16-bit RGB to 16-bit YUV, if this
>> is required?
>
> -vf fillborders doesn't require YUV, so no automatic conversion happens.

You asked for yuv, 16 bit rgb is nothing unusual, see ticket #503 for
a sample with transparency.

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis
In reply to this post by Carl Eugen Hoyos-2

Am 09.03.19 um 10:58 schrieb Carl Eugen Hoyos:
> 2019-03-09 10:44 GMT+01:00, Ulf Zibis <[hidden email]>:
>> Which format should I use with option -f ?
> You should never need the option -f (except when reading
> raw g.729 and similar rare raw formats), what did you try?
When I use
$ ffmpeg -i 16.jpg -vf fillborders=25:25:25:25:mirror 16_mirror-25.jpg
then the output is 8-bit rgb.

>> The files from ticket #503 are only gray,
> Yes.
>
>> and the JPEG's from ticket #855 seem corrupted
> They are not corrupted as you can easily test with FFmpeg.
>
> (The way I understood your question you asked for 16bit
> images that can be read with FFmpeg, not for 16 bit
> images that can be read with other software which is
> much more difficult.)

Thanks again for the hints, I now can use these picts for my testing.

-Ulf

_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Carl Eugen Hoyos-2
2019-03-11 20:05 GMT+01:00, Ulf Zibis <[hidden email]>:
>
> Am 09.03.19 um 10:58 schrieb Carl Eugen Hoyos:
>> 2019-03-09 10:44 GMT+01:00, Ulf Zibis <[hidden email]>:
>>> Which format should I use with option -f ?
>> You should never need the option -f (except when reading
>> raw g.729 and similar rare raw formats), what did you try?
> When I use
> $ ffmpeg -i 16.jpg -vf fillborders=25:25:25:25:mirror 16_mirror-25.jpg
> then the output is 8-bit rgb.

This seems unlikely as our "jpg" encoder does not support
encoding rgb.

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis
In reply to this post by Paul B Mahol

Am 09.03.19 um 09:39 schrieb Paul B Mahol:
> On 3/9/19, Ulf Zibis <[hidden email]> wrote:
>> Another question is, why the fillborders doesn't check if the input is
>> writable, as suggested in doc/writing_filters.txt.
>>
> Because it request writtable frames.

Hey Paul, how is this managed?
I don't see something like "get writable frame".

-Ulf


_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Paul B Mahol
On 3/11/19, Ulf Zibis <[hidden email]> wrote:

>
> Am 09.03.19 um 09:39 schrieb Paul B Mahol:
>> On 3/9/19, Ulf Zibis <[hidden email]> wrote:
>>> Another question is, why the fillborders doesn't check if the input is
>>> writable, as suggested in doc/writing_filters.txt.
>>>
>> Because it request writtable frames.
>
> Hey Paul, how is this managed?
> I don't see something like "get writable frame".

It is at bottom.
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis
In reply to this post by Carl Eugen Hoyos-2

Am 11.03.19 um 20:14 schrieb Carl Eugen Hoyos:

> 2019-03-11 20:05 GMT+01:00, Ulf Zibis <[hidden email]>:
>> Am 09.03.19 um 10:58 schrieb Carl Eugen Hoyos:
>>> 2019-03-09 10:44 GMT+01:00, Ulf Zibis <[hidden email]>:
>>>> Which format should I use with option -f ?
>>> You should never need the option -f (except when reading
>>> raw g.729 and similar rare raw formats), what did you try?
>> When I use
>> $ ffmpeg -i 16.jpg -vf fillborders=25:25:25:25:mirror 16_mirror-25.jpg
>> then the output is 8-bit rgb.
> This seems unlikely as our "jpg" encoder does not support
> encoding rgb.
Oops, I partly was in error. But the output is 8-bit instead 16-bit and
yuv instead y.

What I have to do, if I want the same format for the output, as it is
for the input?

$ ./ffmpeg -i debug/16.jpg -vf fillborders=25:25:25:25:mirror
debug/16_mirror-25.jpg
ffmpeg version N-93301-gf95aee2b72 Copyright (c) 2000-2019 the FFmpeg
developers
  built with gcc 7 (Ubuntu 7.3.0-27ubuntu1~18.04)
  configuration:
  libavutil      56. 26.100 / 56. 26.100
  libavcodec     58. 47.103 / 58. 47.103
  libavformat    58. 26.101 / 58. 26.101
  libavdevice    58.  6.101 / 58.  6.101
  libavfilter     7. 48.100 /  7. 48.100
  libswscale      5.  4.100 /  5.  4.100
  libswresample   3.  4.100 /  3.  4.100
Input #0, image2, from 'debug/16.jpg':
  Duration: 00:00:00.04, start: 0.000000, bitrate: 103852 kb/s
    Stream #0:0: Video: mjpeg (Lossless),
gray16le(bt470bg/unknown/unknown), 640x480 [SAR 96:96 DAR 4:3],
lossless, 25 tbr, 25 tbn, 25 tbc
File 'debug/16_mirror-25.jpg' already exists. Overwrite ? [y/N] y
Stream mapping:
  Stream #0:0 -> #0:0 (mjpeg (native) -> mjpeg (native))
Press [q] to stop, [?] for help
[swscaler @ 0x555de22ba600] deprecated pixel format used, make sure you
did set range correctly
Output #0, image2, to 'debug/16_mirror-25.jpg':
  Metadata:
    encoder         : Lavf58.26.101
    Stream #0:0: Video: mjpeg, yuvj444p(pc), 640x480 [SAR 1:1 DAR 4:3],
q=2-31, 200 kb/s, 25 fps, 25 tbn, 25 tbc
    Metadata:
      encoder         : Lavc58.47.103 mjpeg
    Side data:
      cpb: bitrate max/min/avg: 0/0/200000 buffer size: 0 vbv_delay: -1
frame=    1 fps=0.0 q=6.0 Lsize=N/A time=00:00:00.04 bitrate=N/A
speed=0.448x   
video:33kB audio:0kB subtitle:0kB other streams:0kB global headers:0kB
muxing overhead: unknown

$ mediainfo debug/16.jpg
General
Complete name                            : debug/16.jpg
Format                                   : JPEG
File size                                : 507 KiB

Image
Format                                   : JPEG
Width                                    : 640 pixels
Height                                   : 480 pixels
Color space                              : Y
Bit depth                                : 16 bits
Compression mode                         : Lossy
Stream size                              : 507 KiB (100%)

$ mediainfo debug/16_mirror-25.jpg
General
Complete name                            : debug/16_mirror-25.jpg
Format                                   : JPEG
File size                                : 33.5 KiB

Image
Format                                   : JPEG
Width                                    : 640 pixels
Height                                   : 480 pixels
Color space                              : YUV
Bit depth                                : 8 bits
Compression mode                         : Lossy
Stream size                              : 33.5 KiB (100%)

-Ulf

_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis
In reply to this post by Paul B Mahol

Am 11.03.19 um 20:27 schrieb Paul B Mahol:

> On 3/11/19, Ulf Zibis <[hidden email]> wrote:
>> Am 09.03.19 um 09:39 schrieb Paul B Mahol:
>>> On 3/9/19, Ulf Zibis <[hidden email]> wrote:
>>>> Another question is, why the fillborders doesn't check if the input is
>>>> writable, as suggested in doc/writing_filters.txt.
>>>>
>>> Because it request writtable frames.
>> Hey Paul, how is this managed?
>> I don't see something like "get writable frame".
> It is at bottom.
Do you mean
    .flags         = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC,

?

-Ulf

_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Carl Eugen Hoyos-2
In reply to this post by Ulf Zibis
2019-03-11 20:31 GMT+01:00, Ulf Zibis <[hidden email]>:

>
> Am 11.03.19 um 20:14 schrieb Carl Eugen Hoyos:
>> 2019-03-11 20:05 GMT+01:00, Ulf Zibis <[hidden email]>:
>>> Am 09.03.19 um 10:58 schrieb Carl Eugen Hoyos:
>>>> 2019-03-09 10:44 GMT+01:00, Ulf Zibis <[hidden email]>:
>>>>> Which format should I use with option -f ?
>>>> You should never need the option -f (except when reading
>>>> raw g.729 and similar rare raw formats), what did you try?
>>> When I use
>>> $ ffmpeg -i 16.jpg -vf fillborders=25:25:25:25:mirror 16_mirror-25.jpg
>>> then the output is 8-bit rgb.
>> This seems unlikely as our "jpg" encoder does not support
>> encoding rgb.
> Oops, I partly was in error. But the output is 8-bit instead 16-bit and
> yuv instead y.

Our jpeg encoder only supports 8-bit yuv
No need to post Mediainfo output, all necessary information
is in the FFmpeg console output.

Carl Eugen
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Ulf Zibis

Am 11.03.19 um 20:43 schrieb Carl Eugen Hoyos:
> Our jpeg encoder only supports 8-bit yuv
Ok, that's the reason, thanks.

> No need to post Mediainfo output, all necessary information
> is in the FFmpeg console output.
Maybe for you. ;-)
For me it is not obvious. I guess you mean the info could be read from

gray16le(bt470bg/unknown/unknown) ./. yuvj444p(pc)

-Ulf

_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
Reply | Threaded
Open this post in threaded view
|

Re: Problem with -vf fillborders on 16 bit image

Gyan
In reply to this post by Ulf Zibis


On 12-03-2019 01:04 AM, Ulf Zibis wrote:

> Am 11.03.19 um 20:27 schrieb Paul B Mahol:
>> On 3/11/19, Ulf Zibis <[hidden email]> wrote:
>>> Am 09.03.19 um 09:39 schrieb Paul B Mahol:
>>>> On 3/9/19, Ulf Zibis <[hidden email]> wrote:
>>>>> Another question is, why the fillborders doesn't check if the input is
>>>>> writable, as suggested in doc/writing_filters.txt.
>>>>>
>>>> Because it request writtable frames.
>>> Hey Paul, how is this managed?
>>> I don't see something like "get writable frame".
>> It is at bottom.
> Do you mean
>      .flags         = AVFILTER_FLAG_SUPPORT_TIMELINE_GENERIC,
>

No, in here:

static const AVFilterPad fillborders_inputs[] = {
     {
         .name           = "default",
         .type           = AVMEDIA_TYPE_VIDEO,
         .config_props   = config_input,
         .filter_frame   = filter_frame,
         .needs_writable = 1,
     },
     { NULL }
};


Gyan
_______________________________________________
ffmpeg-user mailing list
[hidden email]
https://ffmpeg.org/mailman/listinfo/ffmpeg-user

To unsubscribe, visit link above, or email
[hidden email] with subject "unsubscribe".
12