362 Commits

Author SHA1 Message Date
Pascal Massimino
f40542508c better token buffer code
(still not finished, but compiles and works ok)

Change-Id: I7002bf8017e31e7af34a53126072b625d23e2589
2012-01-23 18:54:24 -08:00
Pascal Massimino
7f23678da0 fix for LevelCost + little speed-up
the p0 proba was incorrectly accumulated. Merging its contribution into
the LevelCost[] was creating more problems than anything (esp. with trellis)
so let's just not.

Change-Id: I4c07bfee471085df901228d97b20a4d9606ba44e
2012-01-23 09:21:17 -08:00
Pascal Massimino
7107d54483 further speed-up/cleanup of RecordCoeffs() and GetResidualCost()
(note: Incorporated the cost of bin #1 into the LevelCost[])

Change-Id: I6408b2a822efdb97ad6a3a21d380bc7b0da9c715
2012-01-23 07:01:37 -08:00
Pascal Massimino
fd22104022 Introduce Token buffer (unused for now)
Later, will be useful to collect stats and code bitstream in one single pass

Change-Id: Icf866473958b5eff879e219dbb1d7e350f2268ba
2012-01-23 05:57:12 -08:00
Pascal Massimino
5fa148f47c Merge "speed-up GetResidualCost()" 2012-01-23 02:50:58 -08:00
Pascal Massimino
28a9d9b41a speed-up GetResidualCost()
+ misc cosmetics and code polishing

Change-Id: I5830cd2f268d64c072b1cbccc0a4674833875055
2012-01-23 02:36:38 -08:00
Pascal Massimino
11e7dadd96 Merge "misc cosmetics" 2012-01-22 23:52:29 -08:00
Pascal Massimino
378086bd0a misc cosmetics
Change-Id: Iafc73f9fece13b5ca900a8b9c1039268872a181b
2012-01-22 23:20:53 -08:00
Pascal Massimino
d61479f95b add -print_psnr and -print_ssim options to cwebp.
These will report the 7x7-averaged PSNR or SSIM, using the
new internal function WebPPictureDistortion().
This is for information only. These flags have no encoding impact.

+misc opportunistic cosmetics

Change-Id: I64c0a7eca679134d39062e438886274b22bb643f
2012-01-20 07:37:29 -08:00
Pascal Massimino
2e3e8b2ef5 add a WebPCleanupTransparentArea() method
to 'clean up' the fully-transparent area and make it more compressible
new cwebp flags: -alpha_cleanup (off by default, since gain is not 100% guaranteed)

Change-Id: I74d77e1915eee146584cd61c9c1132a41db922eb
2012-01-18 14:01:24 -08:00
James Zern
552c121715 Merge "mux: plug some memory leaks on error" 2012-01-18 13:46:38 -08:00
James Zern
a2a81f7d61 Merge "fix Mach-O shared library build" 2012-01-18 13:16:12 -08:00
James Zern
b3482c4300 Merge "fix gcc-4.0 apple 32-bit build" 2012-01-18 13:15:23 -08:00
James Zern
e4e3ec19ad fix gcc-4.0 apple 32-bit build
gcc-4.0 defines __PIC__ but not __pic__. This leaves the test for
__pic__ should the inverse case exist.
Fixes issue #103; build failing with:
"error: can't find a register in class 'BREG' while reloading 'asm'"

Change-Id: Ia767a733de6ce0294146f9477ff9c46f0ebe13b0
2012-01-18 13:12:45 -08:00
Urvang Joshi
b0d2fecf25 mux: plug some memory leaks on error
Make sure chunk->data and wpi are not leaked by
MuxAddFrameTileInternal() in case of MEMORY_ERROR in ChunkSetNth().

Change-Id: Ie20e84b92f4bdcb7c3b94520f36b20dd2e730545
2012-01-17 18:40:25 -08:00
Pascal Massimino
f0d2c7a76f pass of cosmetics
Change-Id: Id031221b7499e8cfc460c873d0f8124c9bef3aa3
2012-01-17 18:32:27 -08:00
James Zern
b309a6f901 fix Mach-O shared library build
Add extern to kChunks[] in muxi.h.
Fixes:
ld: duplicate symbol _kChunks in .libs/muxinternal.o and .libs/muxedit.o

Change-Id: Ibeb060f7fdec5fe904097a2443f0cda2f7ede884
2012-01-17 16:51:19 -08:00
James Zern
241ddd38e2 doc: delete mux container pdf
The source is now webp-container-spec.txt.

Change-Id: I43debf616b7eec774b5e1586c286974c7332fb48
2012-01-17 12:54:51 -08:00
James Zern
8b1ba272d2 doc: update VP8 decode guide link
The draft is now RFC6386.

Change-Id: Ief019357af5a5de9c6290cafd34db09d984f58aa
2012-01-17 12:52:27 -08:00
James Zern
7e4371c5b3 WebPMuxCreate: fix unchecked malloc
Change-Id: Ic824cdfbede4abe8e5106904dd17d564bb1b32d4
2012-01-15 11:18:23 -08:00
James Zern
eb425586e1 Merge "have makefile.unix clean up src/webp/*~ too" 2012-01-13 14:27:46 -08:00
James Zern
a85c363186 Merge "correct EncodeAlpha documentation" 2012-01-13 14:27:03 -08:00
James Zern
a33842fdb0 Merge "Update webp container spec with alpha filter options." 2012-01-13 14:25:39 -08:00
Urvang Joshi
8d6490dae6 Incremental support for some of the mux APIs.
This CL adds incremental support for WebPMuxCreate() and
WebPMuxGetFeatures()

Change-Id: I312122f5e817d58366e18ae238cb9e2a493811fb
2012-01-13 14:47:53 +05:30
Pascal Massimino
b8375abd08 have makefile.unix clean up src/webp/*~ too
Change-Id: Ida17b92346055b197fe6ccb60a9ac223ddd46f90
2012-01-12 18:49:29 -08:00
James Zern
b5855fc78d correct EncodeAlpha documentation
The paeth filter was removed in 8ca2076

Change-Id: I743502147da4e0b5506e49382311b8f592973d41
2012-01-12 18:37:45 -08:00
Urvang Joshi
dba37fea0e Update webp container spec with alpha filter options.
Change-Id: I032f350c8c698aeb139f7dd148f01a0a3b012b6d
2012-01-12 18:25:00 -08:00
Pascal Massimino
2e74ec8b5f fix compile under MINGW
there was a type conflict around INT32 (in libjpeg).

Patch by nipmarsh
Issue 102: http://code.google.com/p/webp/issues/detail?id=102

Change-Id: I6425b6b9b9861bcaa8bb1adc8a78541e44031a6e
2012-01-12 04:53:30 -08:00
Pascal Massimino
716d1d7f87 fix suboptimal MAX_LEN cut-off limit
MAX_LEN -> max_len
This was sub-optimal at the end of the picture, when there's
less than MAX_LEN bytes left to match.

Change-Id: I5ebe1fca4e7c112dcd34748a082d1c97f95eb099
2012-01-10 07:41:11 -08:00
Pascal Massimino
57cab7b891 Harmonize the alpha-filter predictions at boundary
often reduces compressed size by ~10's of bytes
+ refactored / sped-up the prediction code (gradient: ~30% faster)

Change-Id: I26bd983655dad4f85d5c5ddc20a1980f384c4dd6
2012-01-10 03:21:13 -08:00
pascal massimino
3a9895340f Merge "Fix bug for Alpha in RGBA_4444 color-mode." 2012-01-09 02:59:59 -08:00
Pascal Massimino
8ca2076de1 Introduce a 'fast' alpha mode
.. where only 2 filtering modes are potentially
tried, instead of all of them. This is fast than the exhaustive 'best'
mode, and not much worse.

Options for cwebp are:
 -alpha_filter none
 -alpha_filter fast      (<- default)
 -alpha_filter best      (<- slow)

Change-Id: I8cb90ee11b8f981811e013ea4ad5bf72ba3ea7d4
2012-01-09 02:37:44 -08:00
Vikas Arora
221a06bb16 Fix bug for Alpha in RGBA_4444 color-mode.
Fix bug for Alpha data in RGBA_4444 color-mode.
The Alpha data is required to be clipped [0, 15] while
converting from 8 bits to 4 bits.

Change-Id: I80705d575c35121beb9633a05ec8823435c79586
2012-01-09 14:07:30 +05:30
James Zern
ad1e163a0d cosmetics: normalize copyright headers
Change-Id: I5e2462b101e0447a4f15a1455c07131bc97a52dd
2012-01-06 14:49:06 -08:00
James Zern
c77424d7eb cosmetics: light include cleanup
remove some unused includes

Change-Id: I07fabf707e8eac7f204757e68dab1057fd21aef1
2012-01-06 14:38:34 -08:00
James Zern
9d0e17c9a9 fix msvc build breakage after 252028a
replace inline with WEBP_INLINE; inline is reserved for c++ code
w/visual studio

Change-Id: Ib06c6a12fd786701740d8b260d9aa310ee81fddf
2012-01-05 18:26:40 -08:00
Urvang Joshi
7c4c177c39 Some readability fixes for mux library
Add a #define for UNDEFINED_CHUNK_SIZE and remove a misleading comment.

Change-Id: I5c56a12f1245e647f913c35c6a34d2e5d658415e
2012-01-05 17:53:51 +05:30
pascal massimino
d8a47e66f7 Merge "Add predictive filtering option for Alpha." 2012-01-05 00:33:32 -08:00
Vikas Arora
252028aaac Add predictive filtering option for Alpha.
Add predictive filtering option for Alpha plane.
Valid range for filter option is [0, 5] corresponding to prediction
methods none, horizontal, vertical, gradient & paeth filter.
The prediction method 5 will try all the prediction methods (0 to 4)
and pick the prediction method that gives best compression.

Change-Id: I9244d4a9c5017501a9696c7cec5045f04c16d49b
2012-01-05 13:55:35 +05:30
pascal massimino
9b69be1cdc Merge "Simplify mux library code" 2012-01-05 00:15:48 -08:00
Urvang Joshi
a056170eca Simplify mux library code
Refactor mux code into Read APIs, Set/Delete APIs and internal
objects/utils.

Change-Id: Ia4ce32ec18cf0c1c75de9084fbb28840d46892b4
2012-01-05 12:23:08 +05:30
James Zern
992187a380 improve log2 test
- add check for native log2 to configure
- use a common define (NOT_HAVE_LOG2) to enable use of local library
  version for non-autoconf platforms without their own version,
  currently msvc and android

This uses a negative (NOT_HAVE_) to simplify the ifdef

Change-Id: Id0610eed507f8bb9c5da338918112853d5c8127a
2012-01-04 18:49:37 -08:00
Pascal Massimino
e852f83205 update Android.mk file list
Change-Id: Idffc497931c85582725e65f1d7b7384bcee0fdd5
2011-12-26 02:39:56 -08:00
Pascal Massimino
a90cb2be6e reduce number of copies and mallocs in alpha plane enc/dec
Change-Id: Ice48d2dd403c18870567ee6e1b210b1eb3d5b44f
2011-12-23 01:21:53 -08:00
James Zern
b1662b0530 fix some more type conversion warnings w/MSVC
API & memory related warnings will be dealt with separately

Change-Id: I8a7893baa4d5035c963fb3c16e0bb4eee742e968
2011-12-21 16:07:16 -08:00
Pascal Massimino
223d8c60aa fix some uint64_t -> int conversion warnings with MSC
Change-Id: I5dc314c9d1a3cbda095be3c90886983500104867
2011-12-21 14:43:52 -08:00
James Zern
c1a0437b87 Merge "simplify checks for enabling SSE2 code" 2011-12-20 17:42:59 -08:00
James Zern
f06817aaea simplify checks for enabling SSE2 code
also fixes build issues under vs11 which has a native arm compiler for
windows 8 targets

Change-Id: Id76c2deae9fc9de147d13ad0d34edffcb5a726c4
2011-12-20 17:41:55 -08:00
James Zern
948d4fe94f silence a msvc build warning
implicit double -> float conversion; avoiding sqrtf (c99) for
compatibility

Change-Id: If9b367dde237fc5173c1445cc60b16aa6ad2821c
2011-12-19 23:04:01 -08:00
James Zern
911795490e vwebp: msvc build tweaks
snprintf -> _snprintf / remove double->float warnings

Change-Id: I5dcdc394cf7e2d37205beaf83ed0f80a18ec06ab
2011-12-16 19:30:33 -08:00