Ronald S. Bultje [Tue, 17 Jul 2012 17:37:06 +0000 (10:37 -0700)]
Merge "Fix bug in reference frame counting." into experimental
Tim Terriberry [Tue, 17 Jul 2012 01:47:05 +0000 (18:47 -0700)]
Merge "Add missing mv_{col,row}_{min,max}_sb variables to MACROBLOCK struct." into experimental
Ronald S. Bultje [Tue, 17 Jul 2012 00:49:37 +0000 (17:49 -0700)]
Fix bug in reference frame counting.
vp8_encode_inter_macroblock() is called in both pick_mb_modes() as
well as encode_sb(), thus the number of macroblocks in the counter
were twice as big as actual numbers. This doesn't affect output.
Change-Id: I6de8a996ee44d2f7f2080d8d2177dd7bc6207c93
Daniel Kang [Wed, 11 Jul 2012 00:23:38 +0000 (17:23 -0700)]
Simplify vp8_decode_mb_tokens_8x8.
Change-Id: Icac5b18e5c832681cea866679796a1dd4278cfb8
Daniel Kang [Tue, 10 Jul 2012 16:36:56 +0000 (09:36 -0700)]
Simplify vp8_decode_mb_tokens.
Change-Id: Ib6866e491994f6c7eeb6c2e5ecd1ff54cc187610
Ronald S. Bultje [Mon, 16 Jul 2012 22:25:59 +0000 (15:25 -0700)]
Add missing mv_{col,row}_{min,max}_sb variables to MACROBLOCK struct.
This allows CONFIG_SUPERBLOCKS experiment to almost compile succesfully,
except for the missing pick_sb_modes() function.
Change-Id: Ib2322f2aacdc371e8066f2eb4a8d761c40490b4d
Ronald S. Bultje [Mon, 2 Jul 2012 18:13:18 +0000 (11:13 -0700)]
Use 8x8 transform for all 16x16 intra prediction modes in keyframes.
Also use the 8x8 transform in the RD loop.
Change-Id: If1a25aa15d3f64e05c9a31875e309d8cd5bc812d
Ronald S. Bultje [Tue, 3 Jul 2012 20:57:27 +0000 (13:57 -0700)]
Keep statistics for chosen intra modes in keyframes also.
Change-Id: Ibfecf1fb0fcf51fb349746fec015b1a3e44a4530
John Koleszar [Thu, 12 Jul 2012 16:24:43 +0000 (09:24 -0700)]
Merge snapshot 2 development history
Create a merge between the experimental development history and the
previously published snapshot.
Change-Id: If320df72a6bbefec53833626d08dbc9678be2d2d
John Koleszar [Thu, 12 Jul 2012 16:22:01 +0000 (09:22 -0700)]
Merge snapshot 1 development history
Merge in the git history leading up to the first experimental
snapshot.
Change-Id: Ifbdbd5bdb585864b3f996c562ef38b6306731251
Yaowu Xu [Fri, 29 Jun 2012 00:03:29 +0000 (17:03 -0700)]
change to enable encoder to clamp UV motion vector
This is to avoid a rare encoder/decoder mismatch for MB using SPLITMV
mode. In decoder, the UV mv can be determined to need clamp, but the
flag is never set in encoder motion vector selection process, and the
clamp is not done in encoding in encoder.
Change-Id: I60520d3f790354c7855dadf03f0978ea9b77e2c0
Daniel Kang [Sat, 30 Jun 2012 00:33:36 +0000 (00:33 +0000)]
Merge "Add 8x8 fDCT unit test." into experimental
Christian Duvivier [Fri, 29 Jun 2012 22:05:11 +0000 (15:05 -0700)]
Get rid of some cast.
Change-Id: Iff5ccf2400bcb7b78c7b70f289ea65c982efddd5
Christian Duvivier [Fri, 29 Jun 2012 21:46:11 +0000 (14:46 -0700)]
Fix warnings.
Change-Id: I6dd27a0ea661e6f10a55d7841021b66ae4fc72ae
Daniel Kang [Thu, 28 Jun 2012 23:26:31 +0000 (16:26 -0700)]
Add 8x8 fDCT unit test.
Also factorize ACMRandom to acm_random.h.
Change-Id: I3b6eeb36fcbf7ae6dd3d2892bc40348f5c17982b
Hui Su [Thu, 14 Jun 2012 02:03:31 +0000 (19:03 -0700)]
Add lossless compression mode.
This commit adds lossless compression capability to the experimental
branch. The lossless experiment can be enabled using --enable-lossless
in configure. When the experiment is enabled, the encoder will use
lossless compression mode by command line option --lossless, and the
decoder automatically recognizes a losslessly encoded clip and decodes
accordingly.
To achieve the lossless coding, this commit has changed the following:
1. To encode at lossless mode, encoder forces the use of unit
quantizer, i.e, Q 0, where effective quantization is 1. Encoder also
disables the usage of 8x8 transform and allows only 4x4 transform;
2. At Q 0, the first order 4x4 DCT/IDCT have been switched over
to a pair of forward and inverse Walsh-Hadamard Transform
(http://goo.gl/EIsfy), with proper scaling applied to match the range
of the original 4x4 DCT/IDCT pair;
3. At Q 0, the second order remains to use the previous
walsh-hadamard transform pair. However, to maintain the reversibility
in second order transform at Q 0, scaling down is applied to first
order DC coefficients prior to forward transform, and scaling up is
applied to the second order output prior to quantization. Symmetric
upscaling and downscaling are added around inverse second order
transform;
4. At lossless mode, encoder also disables a number of minor
features to ensure no loss is introduced, these features includes:
a. Trellis quantization optimization
b. Loop filtering
c. Aggressive zero-binning, rounding and zero-bin boosting
d. Mode based zero-bin boosting
Lossless coding test was performed on all clips within the derf set,
to verify that the commit has achieved lossless compression for all
clips. The average compression ratio is around 2.57 to 1.
(http://goo.gl/dEShs)
Change-Id: Ia3aba7dd09df40dd590f93b9aba134defbc64e34
Daniel Kang [Thu, 28 Jun 2012 23:16:00 +0000 (23:16 +0000)]
Merge "Port Yaowu's 4x4 fDCT test." into experimental
Daniel Kang [Wed, 27 Jun 2012 01:11:33 +0000 (18:11 -0700)]
Port Yaowu's 4x4 fDCT test.
Also fix unit testing.
Change-Id: Iacdc6f1ec53388e093cda1c13e4379e83d4a6535
Yaowu Xu [Wed, 27 Jun 2012 23:37:49 +0000 (23:37 +0000)]
Merge "Added Prediction Filter to Mode Selection" into experimental
Adrian Grange [Mon, 25 Jun 2012 23:23:58 +0000 (16:23 -0700)]
Added Prediction Filter to Mode Selection
Added the ability to optionally filter the prediction data
when inter modes are selected (excludes SPLITMV, for now).
The mode selection loop considers both the filtered and
non-filtered prediction data when choosing mode. The filter
can be turned on/off at the frame-level, or signaled for
each MB.
Change-Id: I1b783c71d95a361ab36c761b07e8a6b06bc36822
Yaowu Xu [Tue, 26 Jun 2012 21:15:53 +0000 (21:15 +0000)]
Merge "Enable unit test framework in experimental branch." into experimental
Ronald S. Bultje [Mon, 25 Jun 2012 16:58:09 +0000 (09:58 -0700)]
Enable unit test framework in experimental branch.
Change-Id: I27c98025b2857e2911005a52dd57c46e2524e815
Deb Mukherjee [Wed, 20 Jun 2012 12:07:24 +0000 (05:07 -0700)]
Continued adaptive entropy coding
Incorporates mv_ref, mbsplit and second_mv into the adaptive
entropy framework. The mv_ref framework has been modified from
before.
Adds some clean-ups and fixes.
Results with the adaptive entropy experiment are currently up by
+1.93% on derf; +2.33% std-hd and +1.87% yt-hd.
Fixed a nasty intermittent bug.
Change-Id: I4b1ac9f9483b48432597595195bfec05f31d1e39
Ronald S. Bultje [Mon, 25 Jun 2012 18:23:09 +0000 (11:23 -0700)]
Sign-extend input argument so it can be used in pointer arithmetic.
Change-Id: I6cbd4de96f9dcc783cef170bfd7652f6cbee36a2
Ronald S. Bultje [Wed, 20 Jun 2012 18:26:33 +0000 (18:26 +0000)]
Merge "x86inc: add some more format identifiers for elf file format recognition." into experimental
Ronald S. Bultje [Wed, 20 Jun 2012 18:00:27 +0000 (18:00 +0000)]
Merge "Reset executable flags for configure." into experimental
Ronald S. Bultje [Wed, 20 Jun 2012 16:48:54 +0000 (09:48 -0700)]
x86inc: add some more format identifiers for elf file format recognition.
Change-Id: I3582c64200eed3606a4b57a9f78624ec46041461
Ronald S. Bultje [Tue, 19 Jun 2012 22:34:49 +0000 (15:34 -0700)]
libs.mk: add x86inc.asm to the CODEC_SRCS.
Change-Id: I45ea0369ad489080c2505269855a46ca1d3624f7
Ronald S. Bultje [Tue, 19 Jun 2012 15:43:41 +0000 (15:43 +0000)]
Merge "Add comments to intra prediction probability tables." into experimental
Christian Duvivier [Fri, 15 Jun 2012 01:14:43 +0000 (18:14 -0700)]
Fix a bunch of warnings.
Change-Id: I883b4f63559c1cf66a25016e946b1b1ae98a20da
Ronald S. Bultje [Mon, 18 Jun 2012 21:21:30 +0000 (14:21 -0700)]
Reset executable flags for configure.
This was accidently disabled in
1fe85a35e0451434d183da71e4e85796d11ca868.
Change-Id: I09dbfecfe45b28dec75b27a627e3065f9c7dc8b2
Daniel Kang [Mon, 18 Jun 2012 19:45:34 +0000 (12:45 -0700)]
x86inc: Move x86inc to the correct location.
Change-Id: I6802731a4d15feef5ce62993dc505ded55c40f7e
Ronald S. Bultje [Mon, 18 Jun 2012 16:22:53 +0000 (09:22 -0700)]
Add comments to intra prediction probability tables.
Change-Id: Iec81423e8e6c50acd9e10f69bc99ad0e9058517d
Ronald S. Bultje [Mon, 18 Jun 2012 19:34:51 +0000 (12:34 -0700)]
x86inc.asm: fix symbol mangling and PIC handling on Win64/Mac64.
Change-Id: I97aa175346683184f9430d880593b291a563e04f
Deb Mukherjee [Tue, 5 Jun 2012 22:25:07 +0000 (15:25 -0700)]
Adaptive entropy coding of coefficients, modes, mv.
This patch incorporates adaptive entropy coding of coefficient tokens,
and mode/mv information based on distributions encountered in a frame.
Specifically, there is an initial forward update to the probabilities
in the bitstream as before for coding the symbols in the frame, however
at the end of decoding each frame, the forward update to the
probabilities is reverted and instead the probabilities are updated
towards the actual distributions encountered within the frame.
The amount of update is weighted by the number of hits within each
context.
Results on derf/hd/std-hd are all up by 1.6%.
On derf, the most of the gains come from coefficients, however for the
hd and std-hd sets, the most of the gains come from the mode/mv
information updates.
Change-Id: I708c0e11fdacafee04940fe7ae159ba6844005fd
Daniel Kang [Fri, 15 Jun 2012 17:13:08 +0000 (17:13 +0000)]
Merge "Adds x86inc.asm and update idct/dequant mmx" into experimental
Daniel Kang [Fri, 8 Jun 2012 00:25:54 +0000 (17:25 -0700)]
Adds x86inc.asm and update idct/dequant mmx
Updates idct/dequant mmx assembly to work with vpnext instead of vp8.
Also adds x86inc.asm
Change-Id: I6e147d5e89177ae449271e97e50d082eb11b078e
Yaowu Xu [Thu, 24 May 2012 19:02:12 +0000 (12:02 -0700)]
changed to use a fixed number for update probabity
This commit is to remove two arrays, which contain the probabilities
of how likely each probability in coef_probs table is updated. The
commit changed to use a fixed number "252".
Surprisedly, the overall impact on quality is close to zero, which
basically says the two big static arrays are not helpful at all.
derf: -0.016%, -0.020%
std-hd: 0.000%, -0.013%
yt: -0.022%, +0.007%
yt-hd: -0.038%, +0.034%
Change-Id: Ifee94d28a37dcab4f1d2b994bd5b07575be42b72
Yaowu Xu [Tue, 8 May 2012 19:38:39 +0000 (12:38 -0700)]
Added the ability to accumulate coef stats across encodings
This commit added the ability to accumulate the coef stats across
different encodings using an intermediate binary stats files. The
accumulation happens only the binary stats file exists in current
directory. The encoder needs to be built with "ENTROPY_STATS" to
allow the output. The commit also fixed a few formating issues in
output stats file.
Change-Id: Ib1a41180aa554845cf51e4421a230b128a3a82b4
Paul Wilkins [Tue, 12 Jun 2012 16:29:46 +0000 (16:29 +0000)]
Merge "Merge of further two pass rc clean up and adjustments." into experimental
Paul Wilkins [Tue, 12 Jun 2012 16:19:13 +0000 (16:19 +0000)]
Merge "Key frame and GF sizing changes." into experimental
Paul Wilkins [Wed, 30 May 2012 11:54:34 +0000 (12:54 +0100)]
Merge ... further two pass rc clean up and adjustments.
Changes to calculation of sr_coded_error to include 0,0 case.
Experimental use of sr_coded_error in calculating correction factor
for estimating the allowable Q range.
Reinstated some code needed for calculating section_intra_rating.
Add flash detection in calculation of KF boost
Increased tolerance in testing candidate key frames (needed with
longer motion search as this tends to slightly increase inter %.
Zbin changes for 8x8.
Other minor adjustments, refactoring and bug fixes.
Reinstated some motion break out clauses in boost loop
as their removal hurt a few 50fps clips badly in the std set.
It may be possible to remove them again later if a better way
can be found of preventing overly long gf intervals.
Change-Id: Iee686d0c31072828bb1ccd2bc63f5f1c7c548ea2
Paul Wilkins [Thu, 7 Jun 2012 12:39:20 +0000 (13:39 +0100)]
Key frame and GF sizing changes.
Changes to the equations for kf and gf minQ and to
the boost calculations for kg and gf
Change-Id: I312031c910e6a575334f49075c32f49a8dfff239
Adrian Grange [Thu, 31 May 2012 16:51:54 +0000 (09:51 -0700)]
Fixed bug where invalid pointer is dereferenced
Variables m & mi were being dereferenced when they might
hold invalid values.
The fix is simply to move these dereferences to after the
point at which mb_row and mb_col are tested for validity.
Change-Id: Ib16561efa9792dc469759936189ea379d374ad20
Paul Wilkins [Wed, 30 May 2012 10:29:49 +0000 (11:29 +0100)]
Rate control fix.
This fix addresses some problems with very complex clips like
handling of flashes on clips like crew (which was made worse
by an earlier patch (derf and std-hd)).
Most clips a small effect but some between 1 & 2%
Derf +0.039, +0.211%
YT +0.042, +0.083%
Change-Id: I65fc7c13afc31482040068544dd65b8808f5cb4a
Paul Wilkins [Tue, 29 May 2012 10:06:49 +0000 (10:06 +0000)]
Merge "fixed one more compiling issue with VC" into experimental
Paul Wilkins [Fri, 18 May 2012 13:46:16 +0000 (14:46 +0100)]
Remove "est_max_qcorrection_factor"
Removed the local scaling factor est_max_qcorrection_factor
and related code to simplify estimateq calculation (little effect
anyway)
Cap range of total correction factor.
Slight change to break out case to turn off arf.
Change-Id: I748187737ba93cfadf016f3dfdf8d2741934067f
Yaowu Xu [Sat, 26 May 2012 04:52:09 +0000 (21:52 -0700)]
fixed one more compiling issue with VC
One more compiling issue, only happened in Visual C under a specific
combination of experiments.
Change-Id: Iba0775dc0ea7342b556348a9cdf60a05a59e0241
Yaowu Xu [Fri, 25 May 2012 17:55:47 +0000 (10:55 -0700)]
fixed a number of compiling issues
the commit fixed a number of compiling issues when some epxeriments
are turned on at the same time.
Change-Id: Idb15b215e2d2a7d25f2707f99ef55a34e7301ce7
Paul Wilkins [Fri, 25 May 2012 09:15:27 +0000 (09:15 +0000)]
Merge "Experimental change to two pass prediction decay calculation." into experimental
Yaowu Xu [Wed, 16 May 2012 19:07:46 +0000 (12:07 -0700)]
changed the way that default probs for 8x8 is set.
The commit changed how baseline 8x8 coefficient probabilities are
initialized, to be consistent with the initialization of baseline
4x4 coefficient probabilities.
The commit does not have any effect on compression.
Change-Id: Ifb3902b5dc0b0c2e6dc3aa5d4a6589d528e58355
Paul Wilkins [Fri, 18 May 2012 15:43:58 +0000 (15:43 +0000)]
Merge "Further firstpass.c changes." into experimental
Paul Wilkins [Wed, 16 May 2012 13:49:30 +0000 (14:49 +0100)]
Experimental change to two pass prediction decay calculation.
Remove dependency on amount and speed of motion as this
may not behave well across different image sizes.
Tweak impact of % inter.
Add in experimental adjustment based on relative quality of an
older second reference frame.
Cap range of decay values allowed.
Some small + effect on derf but -ve on yt & hd at this stage.
Change-Id: I390d6f6ebe67a2eb0b834980d0d4650124980d3e
Paul Wilkins [Thu, 17 May 2012 08:37:30 +0000 (08:37 +0000)]
Merge "Move / re-factor some of boost calculation code." into experimental
Ronald S. Bultje [Tue, 15 May 2012 00:39:42 +0000 (17:39 -0700)]
Rewrite reference frame costing in the RD loop.
I now see I didn't write a very long description, so let's do it
here then. We took a pretty big quality hit (0.1-0.2%) from my
recent fix of the inversion of arguments to vp8_cost_bit() in the
RD reference frame costing. I looked into it and basically the
costing prevented us from switching reference frames. This is of
course silly, since each frame codes its own prob_intra_coded, so
using last frame cost indications as a limiting factor can never
be right.
Here, I've rewritten that code to estimate costings based partially
on statistics from progress on current frame encoding. Overall,
this gives us a ~0.2%-0.3% improvement over what we had previously
before my argument-inversion-fix, and thus about ~0.4% over current
git (on derf-set), and a little more (0.5-1.0%) on HD/STD-HD/YT.
Change-Id: I79ebd4ccec4d6edbf0e152d9590d103ba2747775
Paul Wilkins [Tue, 15 May 2012 17:01:44 +0000 (18:01 +0100)]
Further firstpass.c changes.
base the static image test off a measure of 0,0 motion
instead of the decay accumulator value.
Change "transition to still detection" to compare the
decay rate from successive frames.
Minor tweak to the arf extra boost given based on the
number of frames affected.
Removed unused variable mod_err_per_mb_accumulator.
Change-Id: Idd8360083ad409e45f133ce97dd2488259003e64
Deb Mukherjee [Mon, 14 May 2012 23:21:01 +0000 (16:21 -0700)]
Adds new Directional Intra prediction modes.
Adds 6 directional intra predictiom modes for 16x16 and 8x8 blocks.
Change-Id: I25eccc0836f28d8d74922e4e9231568a648b47d1
Paul Wilkins [Tue, 15 May 2012 15:04:48 +0000 (15:04 +0000)]
Merge "Firstpass.c refactoring" into experimental
Paul Wilkins [Tue, 15 May 2012 15:04:14 +0000 (15:04 +0000)]
Merge "Two pass refactoring continued." into experimental
Paul Wilkins [Tue, 15 May 2012 14:44:34 +0000 (14:44 +0000)]
Merge "Two pass rc refactoring." into experimental
Yaowu Xu [Wed, 9 May 2012 16:31:14 +0000 (09:31 -0700)]
Changed to use integer 8x8 dct
The commit added an integer version of 8x8 forward DCT, based on the
orginal forward DCT from VP6. The constants, roundings, and shifts
were adjusted to improve the accuracy. The latest patch has a very
similar accuracy in term of round trip error against the floating
point version.
It should be noted here that the purpose of the patch is to help
encoding speed and facilitate all other experiments. There will be
futher review in combination with inverse DCT before finalization.
configure with "--enable--int_8x8fdct" to use the integer version
Change-Id: I5a4f80507429f0e07cf02a13768ec81cbfddc5bc
Paul Wilkins [Tue, 15 May 2012 14:28:02 +0000 (15:28 +0100)]
Move / re-factor some of boost calculation code.
Some marginal impact due to the fact that it makes use of
arf more likely / stable even in hard sections.
Change-Id: Ic72fda0f63eefc9433914b5d9cd374d515810129
Paul Wilkins [Tue, 15 May 2012 13:58:13 +0000 (14:58 +0100)]
Firstpass.c refactoring
Removed unused function.
Added tentative code to take error score of an older frame
into account when calculating Q range. However, for now
it is disabled pending merging other changes and testing.
Change-Id: Ie89955e70319dac31b79e3b833e3352712a061ec
Paul Wilkins [Tue, 15 May 2012 10:05:16 +0000 (10:05 +0000)]
Merge "First pass overhaul preparatory change." into experimental
Yaowu Xu [Mon, 14 May 2012 19:28:21 +0000 (19:28 +0000)]
Merge "Reversible WHT pair" into experimental
Ronald S. Bultje [Mon, 14 May 2012 19:25:09 +0000 (19:25 +0000)]
Merge "Don't use compound prediction for golden frames based on alt-ref frames." into experimental
Paul Wilkins [Mon, 14 May 2012 14:13:26 +0000 (15:13 +0100)]
Two pass refactoring continued.
Remove testing of whether we estimate that it will be possible
to code an arf at a lower Q than the ambient Q. This adds quite
a bit of extra code and complexity for marginal gain.
Factored out some code relating to ARNR selection to a separate
function as this is likely to be changed / simplified soon.
Change-Id: Ia1cf060405637ef5bbf7018355437be21d12375f
Paul Wilkins [Mon, 14 May 2012 13:54:54 +0000 (14:54 +0100)]
Two pass rc refactoring.
Removed odd *100 >> 4 factor from boost calculations. Not all the
calculations exactly match what was there before so there may be
some minor impact on results.
Some other minor tidying up in regard to coding conventions.
The specific values of factors and thresholds will likely change as
part of subsequent patches.
Change-Id: Id976321484ac02ba50294cf54fafbc17dda85686
Ronald S. Bultje [Sat, 12 May 2012 00:48:20 +0000 (17:48 -0700)]
Don't use compound prediction for golden frames based on alt-ref frames.
These frames can force reference frame (arf), mode (zeromv) and skip,
which means that if we use compound prediction (i.e. arf+last), we
might use a blend of a perfect (arf) and an imperfect (last) predictor,
leading to semi-garbage display and thus a huge drop in SSIM/PSNR (up
to 10dB for some frames I analyzed).
Gives a +0.2% gain on YT.
Change-Id: If1f2b7899ad165684af3808fd379295e82558cbb
Deb Mukherjee [Fri, 11 May 2012 17:10:30 +0000 (17:10 +0000)]
Merge "Improved index remapping for prob updates." into experimental
Paul Wilkins [Fri, 11 May 2012 17:07:33 +0000 (18:07 +0100)]
First pass overhaul preparatory change.
This is the first patch in a series of changes to the first
pass code. (Broken down for ease of testing/merging/review).
This patch introduces a new stats element "sr_coded_error".
This is the coded error recorded vs the second reference
frame (which is updated such that it lags by at least one frame).
No use is made of the new structure in this change so this patch
should have no material effect.
Removed some ifdefs and deprecated code (#if NEW_BOOST).
Removed twopass.gf_decay_rate (not used any more)
Change-Id: I1be672a73017f7c13fd50fb4f99236aa2ed30916
Jim Bankoski [Fri, 11 May 2012 15:08:12 +0000 (08:08 -0700)]
Fix configure issue with unit test add.
Change-Id: I960c6eb81f8d76c958e8af989700447f581a8812
James Berry [Wed, 2 May 2012 21:25:58 +0000 (17:25 -0400)]
add unit test support via google test
adds unit testing via google test
Change-Id: I144b50a976d79251fc5135186a4e0a5051ed0e8c
Yaowu Xu [Thu, 10 May 2012 00:19:16 +0000 (17:19 -0700)]
Reversible WHT pair
This commit changed the forward and the inverse 4x4 Walsh Hadamard
transform to a new pair, where the inverse transform can pefectly
reconstuct the input to forward transform. It also does so without
changing the input and output value range. Even more, it does not
change the complexity of the transforms.
While it was not expected to improve the results of our current test,
it does improve std-hd set by 0.2% on all metrics. No change on derf.
Change-Id: Ie4f23ddd3a0f3c5fbe97fb58399f860031f99337
Deb Mukherjee [Thu, 3 May 2012 09:22:26 +0000 (02:22 -0700)]
Improved index remapping for prob updates.
Also includes some clean ups and refactoring.
Rebased.
Change-Id: I268c97fe325b4881103fe19f41ae818569e7ccf7
Yaowu Xu [Wed, 2 May 2012 20:53:15 +0000 (13:53 -0700)]
a number of fixes to entropy stats collection
1. block types
There are only three types of blocks for 8x8 transformed MBs, i.e. Y
block with DC does not exist for 8x8 transformed MBs as all MB using
8x8 transform have 2nd order haar transform. This commit introduced
a new macro BLOCK_TYPES_8X8 to reflect such fact.
2. context counters
This commit also fixed the mixed of context_counters between 4x4 and
8x8 transformed MBs. The mixed use of the counters leads me to think
the existing the context probabilities were not properly generated
from 8x8 transformed MBs.
3. redundant collecting in recoding
The commit also corrected the code that accumulates entropy stats by
making sure stats only collected for final packing, not during the
recode loop
Change-Id: I029f09f8f60bd0c3240cc392ff5c6d05435e322c
Deb Mukherjee [Wed, 21 Mar 2012 22:22:21 +0000 (15:22 -0700)]
Expanding the coefficient encoding contexts
This patch expands the set of prev contexts used for video coding
from 3 to 4.
There is a small improvement of the order of 0.08% for derf and
0.15% on the HD set. The tests were rerun after the various merges
last week. There are two columns in each test - the first are the
results with the mbskip change, and the second with expanded contexts
added on top of that.
Derf:
http://www.corp.google.com/~debargha/vp8_results/explibvpx_newentropy_expcontext.html
HD:
http://www.corp.google.com/~debargha/vp8_results/explibvpx_hd_newentropy_expcontext.html
Rebased.
Broke up 80 char lines.
Change-Id: I82d2e72d054e530cbf5ce9aa0e6d85c582965675
Yaowu Xu [Wed, 2 May 2012 18:10:25 +0000 (11:10 -0700)]
chagned the decoder band to match the encoder
missed the decoder side in last commit
Change-Id: Ie97f35189e93f78783e3d8072a36eea768beed27
Yaowu Xu [Wed, 2 May 2012 02:54:19 +0000 (19:54 -0700)]
slight adjustment to coef band definition
This commit adjusted slightly the 4x4 coefficents band definition to
better classify coefficients with similar distributions and usages.
It helps derf set about .1%, it is alos slightly positive for std-hd
set, where 4x4 blocks are used less frequently.
The commit also removed a const array not in use.
Change-Id: I78d16905d4036641ec905b0c32c190c1def5b249
Yaowu Xu [Tue, 1 May 2012 19:25:56 +0000 (12:25 -0700)]
Added usage info on command line options
This commit added usage information that was missed in usage output.
Change-Id: If1a85bcc20131314f5d047c6db4e76a40d945e66
Ronald S. Bultje [Mon, 30 Apr 2012 23:32:12 +0000 (16:32 -0700)]
Fix inversion of probability and value in calls to vp8_cost_bit().
Change-Id: I9f1686249ac812f7b9b872eabe3970d1dfb25e56
Deb Mukherjee [Mon, 30 Apr 2012 19:24:22 +0000 (12:24 -0700)]
Turning off filter search for now to improve encode speed.
Change-Id: I87291fb40c745f34c36b067f47abdf69774a812f
Deb Mukherjee [Mon, 30 Apr 2012 18:03:24 +0000 (11:03 -0700)]
Minor cleanup in tokenize.h
Removes a set of spurious declarations that were inadvertently checked
in.
Change-Id: I2f80b6b66d2ec9ea667c810eaf1a6e7d52478c67
Adrian Grange [Fri, 27 Apr 2012 22:09:57 +0000 (22:09 +0000)]
Merge "Removed MV costing from ARNR filtering" into experimental
Adrian Grange [Thu, 26 Apr 2012 17:37:52 +0000 (10:37 -0700)]
Removed MV costing from ARNR filtering
The ARNR filter uses a motion compensated temporal filter,
but the motion estimation implementation accounts for the
cost of the mv in its decision making process. The ARNR
filter uses a dummy cost table initialized to 0 as a way
to ignore the mv costs (which are irrelevant to the filter).
This CL modifies the ARNR filter implementation to so that
the mv costing is ignored without the requirement for
dummy tables.
Change-Id: I0dd9620c3b70682f938b2a70912c11d4d7c9284c
Deb Mukherjee [Mon, 16 Apr 2012 21:53:37 +0000 (14:53 -0700)]
Adds search option for best interpolation filter.
Adds a speed feature to conduct a brute force search among a set of
available interpolation filters for the best one in an RD sense.
There is a gain of 0.4% on derf, 1.0% on Std-HD.
Patch 2: A macro added to determine if the encoder state is reset
for each new filter tried.
Patch 3: rebase, also fixes a bug (decodframe.c) introduced by a
couple of missing function pointer assignements.
Patch 4: rebase.
Change-Id: Ic9ccca9d8c35c6af557449ae867391a2f996cc29
Yaowu Xu [Wed, 25 Apr 2012 15:47:33 +0000 (08:47 -0700)]
Merge QIMODE experiment
This commit merge the QI mode experiment. As the experiment affects
the encoding of intra coding modes on key frame only, the overall
effect of the experiment on encoding tests is insignificant.
Change-Id: I9e4e3933adface88867ad429cee3986e529c511d
Yaowu Xu [Wed, 25 Apr 2012 15:33:30 +0000 (08:33 -0700)]
Merge UVINTRA experiment
The commit merges the UVINTRA experiment and removed the related
macros. The overall effect of the experiment is a small gain (.1%
on derf)
Change-Id: Ia34b3312fb9b5b34c9ba111bf0fa78c6f78ac80b
Deb Mukherjee [Tue, 24 Apr 2012 23:07:18 +0000 (23:07 +0000)]
Merge "Differential encoding of probability updates" into experimental
Ronald S. Bultje [Tue, 24 Apr 2012 18:07:27 +0000 (11:07 -0700)]
Remove unused header files.
Change-Id: I8708358bb37edabcbe5dfc755ed18791d9e143c4
Deb Mukherjee [Thu, 12 Apr 2012 16:24:03 +0000 (09:24 -0700)]
Differential encoding of probability updates
Adds differential encoding of prob updates using a subexponential
code centered around the previous probability value.
Also searches for the most cost-effective update, and breaks
up the coefficient updates into smaller groups.
Small gain on Derf: 0.2%
Change-Id: Ie0071e3dc113e3d0d7ab95b6442bb07a89970030
Ronald S. Bultje [Fri, 20 Apr 2012 21:25:36 +0000 (14:25 -0700)]
Hide some code behind CONFIG_COMP_INTRA_PRED.
Change-Id: I7c0597dede20cc71145c053f76bd99aaf759d144
Yaowu Xu [Fri, 20 Apr 2012 16:14:26 +0000 (09:14 -0700)]
change to allow 8x8 transform always
This commit changed to enable the usage 8x8 transform for all frame
type, all resolution and all quantizer range. This has an overall
benefit .2% to .3% in term of compression, but more importantly,
the difficult clips benefits much more, up to 2% to 3% on clips
like football, harbour and so on.
We observed some weird humps on very high end on a couple of youtube
clips, but have determined the underly cause was the aggressive zbin
having an effect of lowering rate with lower quality, which have
an impact on slide show clips around 60DB.
The commit does not change the association between prediction mode
and transform size.
Change-Id: I33043bdce6207528ae00b4a4b26d8ff63cfea1f4
Yaowu Xu [Tue, 17 Apr 2012 22:14:45 +0000 (15:14 -0700)]
added reset of rate estimates for each mode
This is to prevent the evaluation of a mode from using values left
over from a mode evaluated prior in the loop.
Change-Id: Ife2c6ceb76d2f7365fd262515d3ae48229033c2d
Ronald S. Bultje [Thu, 19 Apr 2012 17:07:14 +0000 (10:07 -0700)]
Fix splitmv/compound prediction when eightpel is enabled.
Change-Id: I9d6083d54e3d478ec20dc6dc48d3f45eb5c7e16b
Ronald S. Bultje [Wed, 18 Apr 2012 20:51:58 +0000 (13:51 -0700)]
Compound prediction for splitmv macroblocks.
Change-Id: I0af3395500b1cb0ed629249eb6636a0c9322cb18
Adrian Grange [Wed, 18 Apr 2012 16:08:55 +0000 (09:08 -0700)]
Added update of mode context pointers in decoder
With the NEWENTROPY experiment enabled encoding certain clips
produced invlid bitstreams, or files that had a high degree
of artefacts.
This was the results of pointers in MACROBLOCKD not being
setup correctly (mode_info_context and prev_mode_info_context).
Change-Id: Ice13e1efa8bd122997d2f8f3f1e761c6c16e0403
Adrian Grange [Wed, 18 Apr 2012 16:21:16 +0000 (16:21 +0000)]
Merge "Added save coding context & modified MV bounds" into experimental
Yaowu Xu [Tue, 17 Apr 2012 18:37:23 +0000 (18:37 +0000)]
Merge "Added intra mode probabilites into coding_context" into experimental
Yaowu Xu [Tue, 17 Apr 2012 16:26:19 +0000 (09:26 -0700)]
Added intra mode probabilites into coding_context
These contexts need to be saved and restored for recode, otherwise
encoder/decoder mismatch happens for some clips (eg._mobcal 720p)
Change-Id: Ic65cfa0bf56ed0472ecab962ce31394d59d344bf