Total Pageviews

Sunday, 22 May 2016


libbpg , 

BPG Image library and utilities

1) Quick introduction

- Edit the Makefile to change the compile options (the default compile
  options should be OK for Linux). Type 'make' to compile and 'make
  install' to install the compiled binaries.

- bpgview: in order to compile it you need to install the SDL and
  SDL_image libraries.

- Emscripten usage: in order to generate the Javascript decoder, you
  must install Emscripten and enable its use in the Makefile.

- An HTML demonstration (with a precompiled Javascript decoder) is
  available in html/index.html (if you use Chrome and want to use
  file:// to access it, launch Chrome with the option

- The BPG file format is specified in doc/bpg_spec.txt.

2) Compilation and Installation Notes

2.1) Linux

  - Edit the Makefile to change the compile options (the default
  compile options should be OK). Type 'make' to compile and 'make
  install' to install the compiled binaries.
  - Use 'make -j N' where N is the number of CPU cores to compile faster.

  - The following packages must be installed: SDL-devel
  SDL_image-devel yasm. It is recommended to use yasm version >= 1.3.0
  to have a faster compilation.
  - Only a 64 bit target is supported because x265 needs it for bit
    depths > 8.

2.2) Windows

  - Only cross-compilation from Linux is supported.

  - The following packages need to be installed: mingw64-gcc
    mingw64-libpng mingw64-libjpeg-turbo mingw64-SDL mingw64-SDL_image
    yasm. It is recommended to use yasm version >= 1.3.0 to have a
    faster compilation.

  - Only a 64 bit target is supported because x265 needs it for bit
    depths > 8.

3) BPG encoder

The BPG command line encoder is 'bpgenc'. It takes JPEG or PNG images
as input.

- Speed: by default bpgenc uses the x265. You can compile the much
  slower but more efficient JCTVC encoder and select it with the '-e
  jctvc' option. With x265 you can select the encoding speed with the
  '-m' option (1 = fast, but larger image, 9 = slower but smaller

- Bit depth: the default bit depth is 8. You can increase it to 10
  ('-b 10' option) to slightly increase the compression ratio. For web
  publishing it is generally not a good idea because the Javascript
  decoder uses more memory. The compiled x265 encoder supports the bit
  depth of 8, 10 and 12. The slower JCTVC encoder can be compiled to
  support higher bit depths (up to 14) by enabling the Makefile

- Lossless compression is supported as a bonus thru the HEVC lossless
  capabilities. Use a PNG input in this case unless you know what you
  do ! In case of a JPEG input, the compression is lossless related to
  the JPEG YCbCr data, not the RGB data. In any case, the bit depth
  should match the one of your picture otherwise the file size
  increases a lot. By default the lossless mode sets the bit depth to
  8 bits. The prefered color space is set to "rgb". Notes: 
    - lossless mode is less tested that the lossy mode but it usually
      gives better results that PNG on photographic images.

    - the JCTVC encoder gives smaller images than the x265 encoder
      with lossless compression.

- There is a small difference of interpretation of the quantizer
  parameter (-q option) between the x265 and JCTVC encoder.

- Color space and chroma format:

    * For JPEG input, the color space of the input image is not
      modified (it is YCbCr, RGB, YCbCrK or CMYK). The chroma is
      subsampled according to the preferred chroma format ('-f'

    * For PNG input, the input image is converted to the preferred
      color space ('-c' option). Its chroma is then subsampled
      according to the preferred chroma format.

    * grayscale images are kept unmodified.

- Premultiplied alpha: by default bpgenc uses non-premultiplied alpha
  to preserve the color components. However, premultiplied alpha
  ('-premul' option) usually gives a better compression at the expense
  of a loss in the color components. This loss is not an issue if the
  image is not edited.

- Animations: with the '-a' option, animations can be encoded from a
  sequence of PNG or JPEG images, indexed from 1 or 0. For example:

  ./bpgenc -a anim%2d.png -fps 25 -loop 0 -o anim.bpg

  generates an animation from anim01.png, anim02.png, etc... The frame
  rate is specified with '-fps' and the number of loops with '-loop'
  (0 = infinite). If a different delay per image is needed as in some
  animated GIFs, a text file can be specified with the '-delayfile'
  option. It contains one number per image giving its duration in
  centiseconds. All durations are rounded to a multiple of '1/fps', so
  it is important to set a consistent frame rate.
  The necessary frames and delay file can be generated from animated
  GIFs with the ImageMagick tools:
  convert -coalesce anim.gif anim%d.png
  identify -format "%T\n" anim.gif > anim.txt
  In order to reduce the file size, the frame rate can be choosen so
  that most frames have a frame period of 1 (hence if anim.txt
  contains only frame durations of 5 centiseconds, then choose a frame
  rate of 20 frames/s).

  As GIFs use paletted colors and 1 bit transparency, it is always
  better to start from the source material (e.g. PNG files) to have
  the best quality.

  A BPG decoder not supporting animations only displays the first

- By default, bpgenc does not copy the metadata. You can copy them
  with the '-keepmetadata' option. For JPEG input, EXIF, ICCP and XMP
  are copied. For PNG input, ICCP is copied.

- Objective comparisons: x265 is tuned by default for SSIM. the JCTVC
  encoder is tuned for PSNR only, not for SSIM, so you should use PSNR
  when making objective comparison with other formats.

4) BPG decoder

The BPG command line decoder is bpgdec. It outputs a PNG or PPM
image. Use a PPM output to get the fastest speed.

- With the '-i' option, you have information about the BPG image (and
no decoded image is output).

- The '-b' option selects the bit depth (8 or 16) of the PNG
  output. It is independent of the internal BPG bit depth.

5) BPG viewer

The BPG image viewer uses the SDL library to display BPG images and
other image formats supported by the SDL_image library. The available
keys are displayed by launching bpgview without parameters. bpgview
supports BPG animations.

6) BPG decoding library

BPG images can be decoded in any program with the libbpg

The API is not considered stable yet so that the library is only
provided as a static one.

Currently there is no similar library for encoding so you should
invoke the bpgenc utility.

7) Javascript decoder

The following Javascript decoders are available, sorted by increasing size:

             > 8 bits   animations
bpgdec8.js   no         no
bpgdec.js    yes        no
bpgdec8a.js  no         yes

The 8 bit only decoders are a little faster and consumes less memory
(16 MB instead of 32 MB by default, you can change the memory
configuration in the Makefile if you want to handle larger images).

The Javascript decoder substitutes all the <img> tags with a source
having a .bpg extension with a <canvas> tag and decodes the BPG image
into it. Stylesheets are supported (the 'id' and 'class' attributes
are preserved). The 'width' and 'height' attributes are supported only
with pixel units.

The image data is downloaded with the XMLHttpRequest object. So the
BPG images and the BPG Javascript decoder must be in the same domain
unless Cross-Origin Resource Sharing is used.

When animations are displayed, all the frames are stored in memory, so
animations with a large number of frames and large resolutions should
be avoided, as with animated GIFs.

asm.js gives an interesting speed boost, so we hope that more browsers
will support this Javascript subset.

8) FFmpeg modifications

- Completed support of chroma_format_idc = 0 (monochrome mode).

- Fixed RDPCM support (intra predictions).

- Reduced memory usage for the SAO loop filter.

- Generated the IDCT coefficients dynamically to reduce the code size.

- Added a 'dynamic bit depth' mode where all the bit depths from 8 to
  14 are supported without code duplication but slower decoding.

- Added a modified SPS header to reduce the size of the BPG decoder
  (an alternate solution is to generate standard VPS and SPS headers
  from the BPG header).

- Added defines to keep only the HEVC intra code and suppress the
  parsing of all the irrelevant NAL units.

- Stripped FFmpeg from all codecs except HEVC and the necessary
  support code.

9) x265 modifications

- Support of monochrome format (some parts not used by BPG may be

- Support of static build.


BPG Image format



BPG (Better Portable Graphics) is a new image format. Its purpose is to replace the JPEG image format when quality or file size is an issue. Its main advantages are:
  • High compression ratio. Files are much smaller than JPEG for similar quality.
  • Supported by most Web browsers with a small Javascript decoder (gzipped size: 56 KB).
  • Based on a subset of the HEVC open video compression standard.
  • Supports the same chroma formats as JPEG (grayscale, YCbCr 4:2:0, 4:2:2, 4:4:4) to reduce the losses during the conversion. An alpha channel is supported. The RGB, YCgCo and CMYK color spaces are also supported.
  • Native support of 8 to 14 bits per channel for a higher dynamic range.
  • Lossless compression is supported.
  • Various metadata (such as EXIF, ICC profile, XMP) can be included.
  • Animation support.


The following archive contains the source code of the bpgencbpgdec and bpgview command line utilities (for Linux) and the associated libbpg library (read the README file in the archive). It also includes the source code of the Javascript decoder.
Binary distribution for Windows (64 bit only):
Unofficial Github mirror.
For Mac users, the BPG utilities are available in the libbpg Homebrew formula.


  • Mozilla did a study of various lossy compressed image formats. HEVC (hence BPG) was a clear winner by a wide margin. BPG files are actually a little smaller than raw HEVC files because the BPG header is smaller than the corresponding HEVC header.
  • BPG natively supports 8 to 14 bits per channel when most other formats use 8 bits (including most of the JPEG implementations and WEBP). It gives a higher dynamic range (which is important for cameras and new displays) and a slightly better compression ratio (because there are less rounding errors in the decoder).
  • BPG uses high quality decimation (10 tap Lanczos filter) and interpolation (7 tap Lanczos filter) to handle the chroma samples in 4:2:2 and 4:2:0 formats.
  • BPG can be supported in hardware with standard HEVC decoders and encoders (it uses a subset of the Main 4:4:4 16 Still Picture Profile, Level 8.5).


Technical information

The specification of the BPG file format is available here.


  • The BPG decoding library uses a modified version of FFmpeg released under the LGPL version 2.1 as HEVC decoder. The BPG decoding library excluding the FFmpeg code is released under the BSD license.
  • The BPG encoder as a whole is released under the GPL version 2 license. The BPG encoder sources excluding x265 are released under the BSD license. The x265 library is released under the GPL version 2 license. The optional JCTVC HEVC reference encoder is released under the BSD license.
  • Some of the HEVC algorithms may be protected by patents in some countries (read the FFmpeg Patent Mini-FAQ for more information). Most devices already include or will include hardware HEVC support, so we suggest to use it if patents are an issue.