Even at high quality setting, the U/V quantizer step is limited
to 4 which can lead to banding on gradient.
This option allows to selectively apply some randomness to
potentially flattened-out U/V blocks and attenuate the banding.
This option is off by default in 'dwebp', but set to -dither 50
by default in 'vwebp'.
Note: depending on the number of blocks selectively dithered,
we can have up to a 10% slow-down in decoding speed it seems.
Change-Id: Icc2446007f33ddacb60b3a80a9e63f2d5ad162de
When '-mixed' option is given, each frame would be heuristically chosen
to be
encoded using lossy or lossless compression.
The heuristic is based on the number of colors in the image:
- If num_colors <= 31, pick lossless compression
- If num_colors >= 194, pick lossy compression
- Otherwise, try both and pick the one that compresses better.
Change-Id: I908c73493ddc38e8db35b7b1959300569e6d3a97
This is to conform to man/gif2webp.1
Earlier, one needed to give both '-kmin 0' and '-kmax 0' for this to
work.
Also, suppress further warnings for kmin = 0 and/or kmax = 0 case.
Change-Id: I6f5eeb609aeffc159d0252a40a5734162f7e4e7d
Specifically:
- Merge OptimizeAndEncodeFrame with WebPFrameCacheAddFrame: they use the same
if-else structure.
- Move maintenance of 'prev_canvas' and 'curr_canvas' to util.
- Move ReduceTransparency() and FlattenPixels() calls to SetFrame(): This is in
preparation for the next patch: which will try try lossless encoding for
each frame, even when '-lossy' option is given.
- Make most methods static inside util.
No changes to output expected.
Change-Id: I1f65af25246665508cb20f0f6e338f9aaba9367b
quite rarely, it gives a different output
then without the HINT, but that's often for
a smaller size (tested with default -m and -m 6)
Change-Id: I51d221ab61f8e007983325031345728e8d80b241
helps during lossless compression.
10% average saving, but that's mostly on what was previously
'difficult' cases, where the gain is ~30-50% actually.
Non-difficult cases are mostly unchanged.
Tested over ~7k random web gifs.
Change-Id: I09db4560e4ab09105d1cad28e6dbf83842eda8e9
We reduce transparency by turning some transparent pixels into
corresponding RGB values from previous canvas.
This improves compression by about 23%.
Change-Id: I02d70a43a1d0906ac09a7e2dc510be3b2d38f593
useful for testing. Not listed in the man or README, since
it's only useful for testing the incremental decoding
(e.g. measuring the timing difference compared to non-incremental)
Change-Id: I8df8046e031d21006242babb5bcac09f8ff9f710
...and make gif2webp and vwebp compile without them.
Makefile.vc still to be updated...
Meanwhile the CL environment variable can be supplemented with
set CL=/DWEBP_HAVE_GL /IC:\opt\freeglut\include
Change-Id: I37a60b8c32aafd125bffa98b6cc9f57c022ebbd0
We use the 'do not blend' option for creating independent frames.
We also mark the already independent frames as 'do not blend'.
This bounds the maximum number of frames that need to be decoded to
decode a given frame, thus leading to a much better decoding performance.
Change-Id: I7cef98af2b53751ec36993fd2bd54f7f4c4aad2b
Doesn't work with WIC
+ redirect some info messages from stdout to stderr
+ fix the error reporting upon output-writing error
Change-Id: I92b8bd7a15e656a3f3cdfbf56299f024e39453f8
Earlier, all lossless images were assumed to contain alpha.
Now, we use the 'alpha_is_used' bit from the VP8L bitstream to determine
the
same.
Detecting an absence of alpha can sometimes lead to much more efficient
rendering, especially for animated images.
Related: refine mux code to read width/height/has_alpha information only
once
per frame/fragment. This avoid frequent calls to VP8(L)GetInfo().
Change-Id: I4e0eef4db7d94425396c7dff6ca5599d5bca8297
This is to ensure that the output WebP image is consistent with original
GIF when displayed against any background color.
Change-Id: I14218848153eb40358aa4ce331b2543d2fc2e86c
from x86_64-w64-mingw32-gcc
examples/wicdec.c: In function ‘ExtractICCP’:
examples/wicdec.c:131:21: warning: format ‘%u’ expects argument of type
‘unsigned int’, but argument 4 has type ‘size_t’ [-Wformat]
Change-Id: I6642dae62265a2276ae9ac96dd8ce6f1e2d37ca5
rather than symlink the webm/vpx terms, use the same header as libvpx to
reference in-tree files
based on the discussion in:
https://codereview.chromium.org/12771026/
Change-Id: Ia3067ecddefaa7ee01550136e00f7b3f086d4af4
They went out of sync some time ago, and are
no longer really required since we have them
buildable from makefile.unix
Change-Id: Ica2dcf5c55f44365598f832f55204d123d7aa601
new option: -blend_alpha 0xrrggbb
also: don't force picture.use_argb value for lossless. Instead,
delay the YUVA<->ARGB conversion till WebPEncode() is called.
This make the blending more accurate when source is ARGB
and lossy compression is used (YUVA).
This has an effect on cropping/rescaling. E.g. for PNG, these
are now done in ARGB colorspace instead of YUV when lossy compression
is used.
Change-Id: I18571f1b1179881737a8dbd23ad0aa8cddae3c6b
When transparent color index and background color index are same,
we should set background color to 0x00ffffff (transparent).
For this, we delay setting the background color until we have read the
first frame.
Change-Id: I443609b9c7697a2b94a66992460cff8465b3c127
buffer the last frame's details to perform DISPOSE_BACKGROUND on the
image's area, rather than the entire canvas.
also fixes transparent backgrounds with animated images
Change-Id: I53e4d70c441e1eeb136f1d01e7c88de4f9ecff53
this would require a PRIuS or similar macro for proper platform
compatibility (Visual Studio for instance would be variants of %lu)
Change-Id: I1af530c7c358c91b845acde1d8c12ef46c2ef746
METADATA_ICCP was renamed to METADATA_ICC in
d8dc72a examples: normalize icc related program arguments
but was merged without rebasing after
0bc4268 cwebp: output metadata statistics
Change-Id: Ie317208488cc851d5d21300591c91cebf5abd4a7
older versions of automake (1.9) it seems would install the headers
regardless of the fact that the library was marked noinst_
this change follows some of the header guidance found here:
http://www.gnu.org/software/automake/manual/automake.html#Headers
Change-Id: I80acc00935097ebf36004e9871574fb9ef09aabf
using token-buffer (that is: slightly more memory. O(output_size))
This change is ON by default. To return to previous behaviour, use
'cwebp -low_memory' or set config.low_memory to true.
Side-effect of this new mode: it forces 1 partition only (which was
default anyway), and makes some statistics about the bitstream
no longer available. cwebp will no longer report 'intra4-coeffs', etc.
This mode also doesn't work (yet) with multi-pass, and -low_memory
is currently forced for multi-pass.
also: reversed the flag: USE_TOKEN_BUFFER -> DISABLE_TOKEN_BUFFER
also: fixed the kAverageBytesPerMB estimate
Change-Id: I4ea80382038d6df4309663e0cb7bd88d9bca9cf1
new option: 'cwebp -mt ...'
new config flag: config.thread_level
(allowed thread_level are 0 or 1 for now. Maybe more later...)
If -mt is activated (and WEBP_USE_THREAD is used for compile), the alpha-compression
will be done in parallel to RGB coding for lossy. Can save quite a bit of latency...
Has no effect for lossless encoding.
Change-Id: I769d0bf90e7380cf99344ad62cd77277f4df5a46
