Commit Graph

3463 Commits

Author SHA1 Message Date
hui su
7ec9552c22 Make gif transparent color to be transparent black
Change it from transparent white to transparent black, which matches
the transparent color assumed in Webp dispose-to-background method.

Also pre-multiply background colors before comparison in anim_diff,
just as what is done with regular pixel values.

Change-Id: I5a790522df21619c666ce499f73e42294ed276f2
(cherry picked from commit 43bd895879)
2016-12-08 15:27:40 -08:00
Pascal Massimino
b3fb8bb602 slightly faster Predictor #11 in NEON
(+some slight modifications on Predictor #12)

Change-Id: Ic2132dcd83d961cd069fa01ca1670e35e35274e2
2016-12-08 07:32:51 -08:00
Vincent Rabaud
9871335fc8 Add a CMake option for WEBP_SWAP_16BIT_CSP.
Change-Id: I79ed6cbce9563bf7ca97fc2e10ec3e75b6b8ab5b
(cherry picked from commit 7b54e26bac)
2016-12-07 23:41:36 -08:00
Vincent Rabaud
0ae32226ce Fix missing cpu-features for Android.
Change-Id: Ief13b7c63b96194970b00a5732aa080f794e7ae6
(cherry picked from commit d2223d8d6c)
2016-12-07 23:41:28 -08:00
James Zern
ab4c8056e0 cpu.cmake: improve webp_check_compiler_flag output
WEBP_HAVE_FLAG_LOCAL -> WEBP_HAVE_FLAG_${WEBP_SIMD_FLAG} this will
include the flag being tested in the output:
-- Performing Test WEBP_HAVE_FLAG_NEON

Change-Id: I1c0a143a857b16e4eb1fcf8b23c176380a5fef29
(cherry picked from commit ee1057e3b1)
2016-12-07 23:41:23 -08:00
Vincent Rabaud
eec5fa3a95 Provide support for CMake on Android studio 2.2.
Change-Id: Ib3508c20a220b29d42f963cf154cd9ab5236ed6e
(cherry picked from commit a80e8cfdf2)
2016-12-07 23:41:17 -08:00
Vincent Rabaud
004d569086 Split the main CMake file.
This will well isolate contributions for original code,
generated code and SIMD (especially for Android).

Change-Id: Ie47664decc7f43c2f57260a72cab951c347281a7
(cherry picked from commit 6c62841076)
2016-12-07 23:41:08 -08:00
James Zern
4fe5d588bf Android.mk: use -fvisibility=hidden
brings the final libwebp.so size down 16/20K with arm64/armv7 builds
using ndk-r13

Change-Id: I20d8aba61d6b692b0fc32f4b271e2f9872f03c28
(cherry picked from commit de568abfdb)
2016-12-07 18:34:46 -08:00
James Zern
bd63a31aab vwebp: ensure setenv() is available in stdlib.h
quiets a -Wimplicit-function-declaration with some configurations of gcc
(-std=c99).
_POSIX_C_SOURCE is preferred over _BSD_SOURCE with newer versions of
glibc

Change-Id: I378bffb13ba52ff5c4bad1433090dcc387e5d507
(cherry picked from commit bfab894739)
2016-12-07 18:30:54 -08:00
Pascal Massimino
363a568131 vwebp: handle window resizing properly
The image is scaled to fit the whole viewport.
Avoid some oddities with offsets, etc.

removes some TODO.

Change-Id: I52fae9ca80a2feed234f32261c7f6358d7594e21
(cherry picked from commit 9310d19258)
2016-12-07 18:30:49 -08:00
James Zern
a0d2753fcb lower WEBP_MAX_ALLOCABLE_MEMORY default
restrict to 2^34 for 64-bit targets, < 2^32 for 32-bit

Change-Id: Iff4ce40ae2c3c7fc119f018c2128dbe8f744341f
(cherry picked from commit b8384b53d6)
2016-12-07 18:30:44 -08:00
Pascal Massimino
31fe11a57a fix infinite loop in case of PARTITION0 overflow
max_i4_header_bits_ could drop to zero for difficult image and trigger
a loop. Surprisingly, StatLoop() didn't have this bug.

Change-Id: Idc0f9eadef30a2b2f02041b994f25def30901e36
(cherry picked from commit 21e7537abe)
2016-12-07 18:30:39 -08:00
hui su
532215dd29 Change the rule of picking UV mode in MBAnalyzeBestUVMode()
Pick the mode with the smallest alpha.
It only affects m0, in which case the mode decision is not re-examined
later in VP8Decimate(). Tests on some natural content png images show
PSNR increase as well as visual quality improvement.

Change-Id: Iea997e718cd7477160fa05eb7cfb35f4cec2fa9a
(cherry picked from commit 1377ac2ec1)
2016-12-07 18:30:33 -08:00
James Zern
9c75dbd39c cwebp.1: improve some grammar
Change-Id: Id849d7e0d7573f5b8d3b2e807d95e9c628f03b1e
(cherry picked from commit 5b46f7fc80)
2016-12-07 18:30:28 -08:00
Pascal Massimino
af2e05cbdf vwebp: Clear previous frame when a key triggers a redraw
otherwise, transparent areas were accumulating.

Change-Id: I066a96a2bcf0cac750b3df0c02229542b1ed3473
(cherry picked from commit c0a27fd2af)
2016-12-07 18:30:23 -08:00
Hui Su
26ffa2962b Add descriptions of default configuration in help info.
Change-Id: I43188fab5f57bd45aa3e564df52e36cc37b1bb2f
(cherry picked from commit 74f6f9e793)
2016-12-07 18:30:17 -08:00
hui su
7416280d75 Fix an unsigned integer overflow error in enc/cost.h
Change-Id: I9774b59c417c185f09a61a115364b9642976a100
(cherry picked from commit 0b2c58a91c)
2016-12-07 18:29:51 -08:00
hui su
13cf1d2e41 Do token recording and counting in a single loop
Change-Id: I8afd3c486b210bd67888de03e91dde7f78276f89
(cherry picked from commit 0c0fb83211)
2016-12-07 18:29:44 -08:00
hui su
eb9a4b97c5 Reset segment id if we decide not to update segment map
This avoids potential encoder and decoder mismatch.

Change-Id: I5282d3e168afc6193033ad3fce8fbc35618ab2f5
(cherry picked from commit 386e4ba2f0)
2016-12-07 18:25:06 -08:00
James Zern
42ebe3b783 configure: fix NEON flag detection under gcc 6
use a compile check on a separate file to avoid assuming using
arm_neon.h is safe to use without flags when just the file itself is
self-contained with GCC target pragmas.

BUG=webp:313

Change-Id: I48f92ae3e6e4a9468ea5b937c80a89ee40b2dcfd
(cherry picked from commit 0104d730bf)
2016-12-07 18:23:55 -08:00
Pascal Massimino
76ebbfff28 NEON: implement predictor #13
~5-7% faster

Change-Id: I3361b0bbc978f3721168db15778a67337309c18a
2016-12-07 14:58:49 -08:00
Vincent Rabaud
95b12a08ae Merge "Revert Average3 and Average4" 2016-12-07 15:38:56 +00:00
Vincent Rabaud
54ab2e758f Revert Average3 and Average4
Average3 created a slowdown of 1-2% in lossless decoding.
Average4 created a slowdown of 2-3% in lossless decoding.

Change-Id: Ic2e62cdd83fc897887ec2bf41ea7cadbada84fe5
2016-12-07 15:32:33 +01:00
Pascal Massimino
fe12330c81 3-5% faster Predictor #5, #6, #7 and #10 for NEON
Change-Id: Ica48c7088d4384f0888dd171a47e68ebd25729b2
2016-12-07 15:25:33 +01:00
Pascal Massimino
fbfb3bef7b ~2% faster predictor #10 for NEON
Change-Id: Icd9cff90c227d702c3ba319131996c5475094520
2016-12-06 13:47:35 +00:00
Pascal Massimino
d4b7d801db lossless_sse2: use the local functions
...instead of the pointers stored in the array.
Should be faster (inlined) and safer.

Also: suffix explicitly the functions with _SSE2

Change-Id: Ie7de4b8876caea15067fdbe44abfedd72b299a90
2016-12-06 14:20:41 +01:00
Vincent Rabaud
a5e3b22574 Lossless decoder SSE2 improvements.
Change-Id: Ia901014ac63156a2e278b81e035256c30bdf8706
2016-12-06 13:45:09 +01:00
Pascal Massimino
58a1f124c2 ~2% faster predictor #12 in NEON.
Change-Id: I6772bb865d0f72720a65561eb55028e538df236d
2016-12-06 10:24:27 +01:00
Pascal Massimino
906c3b6392 Merge "Implement lossless transforms in NEON." 2016-12-03 16:55:14 +00:00
Vincent Rabaud
d23abe4e9f Implement lossless transforms in NEON.
Change-Id: I2172b1a763eb9dfe25d2b9bf1fb6501d7e192e55
2016-12-03 11:20:22 +00:00
Vincent Rabaud
2e6cb6f34e Give more flexibility to the predictor generating macro.
Change-Id: Ia651afa8322cb5c5ae87128340d05245c0f6a900
2016-12-02 12:33:12 -08:00
Vincent Rabaud
28e0bb7088 Merge "Fix race condition in multi-threading initialization." 2016-12-02 17:45:10 +00:00
Vincent Rabaud
647045305a Fix race condition in multi-threading initialization.
Before, a first thread could enter VP8LDspInitSSE2, set
VP8LPredictorsAdd to an SSE2 version BEFORE another thread
would do the memcpy from VP8LPredictorsAdd to VP8LPredictorsAdd_C
thus leading to a C version actually being the SSE2 one (which
would then create an infinite recursion in the SSE2 predictors
at execution).

Change-Id: I224f4ceab31d38f77a1375a7e2636a6014080e3a
2016-12-02 18:28:57 +01:00
Pascal Massimino
bded7848ea img2webp: fix default -lossless value and use pic.argb=1
Change-Id: I0e5350928c1e58e0901303ee979fb4587f25d6bc
2016-12-02 13:19:02 +01:00
Pascal Massimino
0e61a5134a Merge "img2webp: convert a sequence of images to an animated webp" 2016-12-02 11:59:26 +00:00
Hui Su
1cc79e92ac AnimEncoder: Correctly skip a frame when sub-rectangle is empty.
Change-Id: I0d288bd9561b48cf5a1eae92a1b7106ba44c664e
2016-12-02 11:50:13 +01:00
Pascal Massimino
03f40955a3 img2webp: convert a sequence of images to an animated webp
Usage:

  img2webp [file-level options] [image files...] [per-frame options...]

File-level options (only used at the start of compression):
 -min_size ............ minimize size
 -loop <int> .......... loop count (default: 0, = infinite loop)
 -kmax <int> .......... maximum number of frame between key-frames
                        (0=only keyframes)
 -kmin <int> .......... minimum number of frame between key-frames
                        (0=disable key-frames altogether)
 -mixed ............... use mixed lossy/lossless automatic mode
 -v ................... verbose mode
 -h ................... this help

Per-frame options (only used for subsequent images input):
 -d <int> ............. frame duration in ms (default: 100)
 -lossless  ........... use lossless mode (default)
 -lossy ... ........... use lossy mode
 -q <float> ........... quality
 -m <int> ............. method to use

example: img2webp -loop 2 in0.png -lossy in1.jpg
                  -d 80 in2.tiff -o out.webp

Change-Id: I23771b90eaf0660f420d7ffd304e704155386286
2016-12-02 11:44:17 +01:00
Pascal Massimino
ea72cd60cb add missing 'extern' keyword for predictor dcl
Change-Id: Ibf3db9b6dae91e53524c31cdfccf4678b3fa1135
2016-12-01 08:15:14 +01:00
Vincent Rabaud
67879e6d48 SSE implementation of decoding predictors.
Change-Id: I5c9ae63afc98013cb45ce8a91f051203ac68402c
2016-11-30 12:00:07 +01:00
Pascal Massimino
34aee99026 Merge "vwebp: make 'd' key toggle the debugging of fragments" 2016-11-29 12:36:34 +00:00
Vincent Rabaud
a41296aef5 Fix potentially uninitialized value.
Change-Id: I721695e22474992db3094942b1ad4754ae7c0a02
2016-11-29 13:19:32 +01:00
Pascal Massimino
c85adb33d2 vwebp: make 'd' key toggle the debugging of fragments
it actually disables the disposal / blending method
and just displays the raw delta values.
Useful for debugging.
TODO: Outline the refreshed area with a drawn rectangle?

Change-Id: I6f8cddd0aad8b953cff78a693ae7e8c31def010c
2016-11-28 19:47:23 +00:00
Vincent Rabaud
4239a1489c Make the lossless predictors work on a batch of pixels.
Change-Id: Ieaee34f1f97c375b9e97ef7e9df60aed353dffa1
2016-11-28 17:12:10 +01:00
Pascal Massimino
bc18ebad2e fix extra 'const's in signatures
Change-Id: Ie433d0defbc0c6feae2eb2f11e70082f1affada8
2016-11-25 09:45:52 +01:00
Vincent Rabaud
71e2f5cadf Remove memcpy in lossless decoding.
Change-Id: Iba694b306486d67764e2fc5576c98a974c9b886c
2016-11-24 17:45:24 +01:00
Vincent Rabaud
7474d46e45 Do not use a register array in SSE.
Change-Id: I79cf95bdac1164fc4de899828e9380c23df8d141
2016-11-24 13:06:44 +01:00
Owen Rodley
67748b41db Improve latency of FTransform2.
Benchmarks from vrabaud@:
8BIT/GRAY                corpus speed: faster: -4.3 % , corpus size: unchanged
skal/sources_png_skal    corpus speed: faster: -5.2 % , corpus size: unchanged
images/png_rgb           corpus speed: faster: -5.1 % , corpus size: unchanged
images/lpcb              corpus speed: unchanged, corpus size: unchanged
images/png_big           corpus speed: faster: -1.7 % , corpus size: unchanged
images/png_doc           corpus speed: unchanged, corpus size: unchanged
images/png_1bit          corpus speed: faster: -1.2 % , corpus size: unchanged
images/jpeg_small        corpus speed: unchanged, corpus size: unchanged
images/icip_core1        corpus speed: unchanged, corpus size: unchanged
images/png_gray          corpus speed: faster: -2.5 % , corpus size: unchanged
images/jpeg_high_quality corpus speed: faster: -4.0 % , corpus size: unchanged
images/jpeg              corpus speed: faster: -2.3 % , corpus size: unchanged
images/png_translucent   corpus speed: faster: -2.8 % , corpus size: unchanged
images/gif               corpus speed: faster: -1.4 % , corpus size: unchanged
images/png_opaque        corpus speed: faster: -2.8 % , corpus size: unchanged
images/png_rgb_opaque    corpus speed: unchanged, corpus size: unchanged
images/png_indexed       corpus speed: faster: -2.0 % , corpus size: unchanged
images/all               corpus speed: faster: -1.5 % , corpus size: unchanged
images/png_small         corpus speed: unchanged, corpus size: unchanged
images/png               corpus speed: unchanged, corpus size: unchanged
images/gif_still         corpus speed: faster: -1.6 % , corpus size: unchanged

Change-Id: I69fe11baa188c5d32cbc77a84b8c0deae13d792b
2016-11-24 07:09:50 +00:00
Vincent Rabaud
16951b1905 Merge "Provide an SSE implementation of ConvertBGRAToRGB" 2016-11-23 16:37:35 +00:00
Vincent Rabaud
6540cd0eeb Provide an SSE implementation of ConvertBGRAToRGB
Change-Id: Ida11b079077a47fe3b92754f08aa30d81c301fcf
2016-11-23 16:25:51 +01:00
James Zern
de568abfdb Android.mk: use -fvisibility=hidden
brings the final libwebp.so size down 16/20K with arm64/armv7 builds
using ndk-r13

Change-Id: I20d8aba61d6b692b0fc32f4b271e2f9872f03c28
2016-11-18 19:24:09 -08:00