How to fix: ffmpeg: Picture size 32768x16384 is invalid?

Question

I need to use ffmpeg with very large images (32768x16384). When I try to read such an image I get the error:

Picture size 32768x16384 is invalid

It it does not matter what operation I want to do on the image (like scaling, v360 cutout etc.) it already fails at the read-in stage. I tried JPG and PNG as input formats.

Full log with demonstration operation (scale 32k to 16k)

E:\XXX\Pipeline>"E:\XXX\Pipeline\bin\ffmpeg.exe" -y -i E:\XXX\Pipeline\calibration_sphere32k.jpg -vf scale=16384:-2 "calibration_sphere32k_scaled.jpg"
ffmpeg version N-94664-g0821bc4eee Copyright (c) 2000-2019 the FFmpeg developers
  built with gcc 9.1.1 (GCC) 20190807
  configuration: --enable-gpl --enable-version3 --enable-sdl2 --enable-fontconfig --enable-gnutls --enable-iconv --enable-libass --enable-libdav1d --enable-libbluray --enable-libfreetype --enable-libmp3lame --enable-libopencore-amrnb --enable-libopencore-amrwb --enable-libopenjpeg --enable-libopus --enable-libshine --enable-libsnappy --enable-libsoxr --enable-libtheora --enable-libtwolame --enable-libvpx --enable-libwavpack --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxml2 --enable-libzimg --enable-lzma --enable-zlib --enable-gmp --enable-libvidstab --enable-libvorbis --enable-libvo-amrwbenc --enable-libmysofa --enable-libspeex --enable-libxvid --enable-libaom --enable-libmfx --enable-amf --enable-ffnvcodec --enable-cuvid --enable-d3d11va --enable-nvenc --enable-nvdec --enable-dxva2 --enable-avisynth --enable-libopenmpt
  libavutil      56. 33.100 / 56. 33.100
  libavcodec     58. 55.101 / 58. 55.101
  libavformat    58. 31.104 / 58. 31.104
  libavdevice    58.  9.100 / 58.  9.100
  libavfilter     7. 58.101 /  7. 58.101
  libswscale      5.  6.100 /  5.  6.100
  libswresample   3.  6.100 /  3.  6.100
  libpostproc    55.  6.100 / 55.  6.100
[mjpeg @ 000001eefc88ab80] [IMGUTILS @ 000000acfa1fef30] Picture size 32768x16384 is invalid
[image2 @ 000001eefc878740] decoding for stream 0 failed
[image2 @ 000001eefc878740] Could not find codec parameters for stream 0 (Video: mjpeg (Baseline), none(bt470bg/unknown/unknown)): unspecified size
Consider increasing the value for the 'analyzeduration' and 'probesize' options
Input #0, image2, from 'E:\XXX\Pipeline\calibration_sphere32k.jpg':
  Duration: 00:00:00.04, start: 0.000000, bitrate: 20939328 kb/s
    Stream #0:0: Video: mjpeg (Baseline), none(bt470bg/unknown/unknown), 25 tbr, 25 tbn, 25 tbc
Stream mapping:
  Stream #0:0 -> #0:0 (mjpeg (native) -> mjpeg (native))
Press [q] to stop, [?] for help
[mjpeg @ 000001eefc88d3c0] [IMGUTILS @ 000000acfa1ff050] Picture size 32768x16384 is invalid
Error while decoding stream #0:0: Invalid data found when processing input
Cannot determine format of input stream 0:0 after EOF
Error marking filters as finished
Conversion failed!

EDIT Found out, that the maximum horizontal resolution is: 22976. This is a very odd number.

Answer 1

The general limiting rule is ((Width * 8) + 1024)*(Height + 128) < INT_MAX. The rationale being, that all pixels of an image plane should be accessible using a signed integer index.

On a machine with 32-bit integers, INT MAX is 2147483647.

So, ((32768 * 8) + 1024)*(16384 + 128) = 4345430016 which is > 2147483647.

Now, it's possible there is a bug with the image size check, in that the factor 8 applied to the width is prematurely applied in most cases and can be skipped. You can raise a ticket at trac.ffmpeg.org if you want to explore that possibility. Else, you'll need to use another tool (maybe Imagemagick).