Updates the common convoloution code to support blocks larger than
16x16, and rectangular blocks. This uncovered a bug in the SSSE3
filtering routines due to the order of application of saturation.
This commit fixes that bug, adjusts the unit test to bias its
random values towards the extremes, and adds a test to ensure that
all filters conform to the expected pairwise addition structure.
Change-Id: I81f69668b1de0de5a8ed43f0643845641525c8f0
Moving all the probability updates after frame context selection.
This makes it clean and simple to store all the probs in single
struct that can be sent to hardware codec.
Change-Id: I2ec3de81adbd468d8ef34a914caae80a18c3ef56
List of moved functions: vp9_decode_uniform, vp9_decode_term_subexp,
vo9_inv_recenter_nonneg, vp9_decode_unsigned_max.
Change-Id: Ib518beb90b791690c5c93de17b8bdbf560033b41
Adds RD integration for 32x16, 16x32, 64x32 and 32x64 rectangular blocks.
Derf almost +0.6%, HD a little over +1.0%, STDHD +1.3%.
Change-Id: Id651fdb6a655fdbb5c47009757e63317acfb88a5
Enable recursive partition information coding from SB64X64 down to
MB16X16. The bit-stream syntax is now supporting rectangular block
sizes. It starts from SB64X64 and recursively describes the partition
type of the current block. If the partition type is PARTITION_NONE,
the block is coded as a single unit; if it is PARTITION_HORZ or
PARTITION_VERT, the block is segmented into two independently coded
rectangular units, with no further partition needed; otherwise, the
block is segmented into 4 square blocks. i.e., PARTITION_SPLIT case,
each can be potentially further partitioned.
Forward adaptive probability modeling is used for the partition
information coding, conditioned on the current block size.
Change-Id: I499365fb547839d555498e3bcc0387d8a3587d87
This function is now called from configures the ARNR
filter so it belongs with the other temporal filter
functions.
Change-Id: I64211875918364b5b8edfb97743e573c6def1663
Normalization of the frame boost value was being done
when it reached the value 1028. The intention was to
keep to a range of 10 bits, so it should have been
clipped above 1023.
Change-Id: I0afdddc1d2eb9e7822ec4578903cbe6ec0b33b91
This flag was added to VP8 to allow a mode where MB-level skipping
was not allowed, saving a bit per mb. It was never used in practice,
and hasn't been tested in VP9, so remove it.
Change-Id: Id450ec6904c6d06c1919508e7efc52d05cde5631
tx_type == DCT_DCT check is an implementation detail of iht_add. Also
adding dequant_add_y function with explicit DCT_DCT check inside.
Change-Id: Ia3cb0225601752cdef0ff6f0acd3a09d9dbd8938
In decoder, the scaling calculation, such as (mv * x_num / x_den),
is fairly time-consuming. In this patch, we check if the scaling
happens or not at frame level, and then decide which function to
call to skip scaling calculation when no scaling is needed. Tests
showed a 3% decoder performance gain.
Change-Id: I270901dd0331048e50368cfd51ce273dd82b8733
This is the first CL with vp9_reader changes. All another macro
definitions will be replaced after.
Change-Id: I1c6bd9c9a612ec1663d484d6adb4fb720af54063
Static threshold results slightly up (+0.1% on derf), probably b/c
we now take the filter (sharp/lowpass) into account for the breakout
decision.
Change-Id: I9f597601da434205142afd05f32690e7ba8fd690
This is work-in-progress, it implements multiple ARF
encoding behind an experimental flag.
It adds the ability to insert multiple ARF frames into a
single ARF group. This patch implements the reordering
of the coded frames, and implements a fixed-length coding
pattern. It applies a fixed quantizer strategy based on
where the frame is in the coding sequence.
Further work to modify the rate control strategy is
ongoing and will be submitted via a set of future patches.
In this first step, each ARF group is recursively
bisected and an ARF frame added at that position in the
sequence. The recursion continues until ARF frames are
within MIN_GF_INTERVAL frames.
The code sits behind the "multiple-arf" experimental
flag ("CONFIG_MULTIPLE_ARF"). The experimental flag
"oneshotq" ("CONFIG_ONESHOTQ") also needs to be enabled
for this patch to work correctly.
Change-Id: Ie473b05ebb43ac473c0cfb659b2b8042823085e2
Combine superblock inter predictors into a unified function that
allows configurable block width and height. The inter predictions
of block sizes smaller than 16x16 are handled differently. To be
continued on merging them later.
Change-Id: I14075959dd5e221f00c205c99ca35c1c31ef728e
To match the order of directional intra prediction modes for larger
blocks, also renamed the i4x4 prediction modes to mirror the larger
variants.
Change-Id: I77cea4d0add6c7758460bf9c7a2fe59aca601f0b
The probabilities derived from these statistics are used in bitstream
writing; therefore, we should only do this when we actually decide to
use macroblock coding (over superblock coding). Derf gains +0.15%.
Change-Id: I196814c070a7c79889590658ce10a6eb07454389
The intra predictor supports configurable block sizes. It can handle
intra prediction down to 4x4 sizes, when enabled in BLOCK_SIZE_TYPE.
Change-Id: I7399ec2512393aa98aadda9813ca0c83e19af854
Rename pick_mb_modes to pick_mb_mode, since it now handles only a
single macroblock. This is consistent with pick_sb_mode handling a
single non-macroblock.
Change-Id: I896fdfa06436b2d8c24d6474718cc74420df6b3b
This patch changes the default with the modecoefprob expt
to use mode-based forward updates with one-node pegged
modeling.
The maximum difference with fully trained tables is now
less that 0.1%.
Change-Id: I06b44322e10c6703f93f3c1d48d973b1136a0618
This patch will use the dest buffer instead of the
predictor buffer. This will allow us in future commits
to remove the extra mem copy that occurs in the dequant
functions when eob == 0. We should also be able to remove
extra params that are passed into the dequant functions.
Change-Id: I7241bc1ab797a430418b1f3a95b5476db7455f6a
More specifically, remove vp9_quantize_mb*, vp9_optimize_mb*,
vp9_inverse_transform_mb* and vp9_transform_mb*. Instead, use the
generic _sb* functions that take a size argument, and call them with
BLOCK_SIZE_MB16X16.
Change-Id: I33024afea95d3a23ffbc1df7da426e4645110f29
With these fixed, the codec produces identical results regardless of
what literal values are used for the enum members in BLOCK_SIZE_*.
Change-Id: I26db8e08019b58ba432af1f0950ebe6b0eb4ad8c
Merge various super_block_yrd and super_block_uvrd versions into one
common function that works for all sizes. Make transform size selection
size-agnostic also. This fixes a slight bug in the intra UV superblock
code where it used the wrong transform size for txsz > 8x8, and stores
the txsz selection for superblocks properly (instead of forgetting it).
Lastly, it removes the trellis search that was done for 16x16 intra
predictors, since trellis is relatively expensive and should thus only
be done after RD mode selection.
Gives basically identical results on derf (+0.009%).
Change-Id: If4485c6f0a0fe4038b3172f7a238477c35a6f8d3
The unified dequantization, inverse transform, and adding functions
support rectangular block sizes. Also separate the operations on
luma and chroma components, in the consideration of the txfm_size
for uv components in rectangular block sizes.
Change-Id: I2a13246b2a9086b37d575d346070990d854cc110
The strategy to run fast loop filter picking for encoder speed-up
should be revisited at a later stage.
Change-Id: I3b75e06d767cff41be952a42e63b3292f4eab996
Merge sb32x32 and sb64x64 functions; allow for rectangular sizes. Code
gives identical encoder results before and after. There are a few
macros for rectangular block sizes under the sbsegment experiment; this
experiment is not yet functional and should not yet be used.
Change-Id: I71f93b5d2a1596e99a6f01f29c3f0a456694d728
Restructure the code to avoid the majority of per-block-size
switches, code duplication, etc. All block types (mb/sb32/sb64)
can be handled by the same code.
Change-Id: I4022718d66e31a15a7074e43f3b98cd0a5124ea7
Removing several commented code blocks, using uint32_t and uint8_t types,
removing redundant code.
Change-Id: Ifc5cc9863897925ea2a7cab4f7309ccf28d80bfe
Clamp only the motion vectors inferred from neighboring reference
macroblocks. The motion vectors obtained through motion search in
NEWMV mode are constrained during the search process, which allows
a relatively larger referencing region than the inferred mvs.
Hence further clamping the best mv provided by the motion search may
affect the efficacy of NEWMV mode.
Synchronized the decoding process. The decoded mvs in NEWMV modes
should be guaranteed to fit in the effective range. Put a mv range
clamping function there for security purpose.
This improves the coding performance of high motion sequences, e.g.,
derf set:
foreman 0.233%
husky 0.175%
icd 0.135%
mother_daughter 0.337%
pamphlet 0.561%
stdhd set:
blue_sky 0.408%
city 0.455%
also saw sunflower goes down by -0.469%.
Change-Id: I3fcbba669e56dab779857a8126a91b926e899cb5
Start grouping data per-plane, as part of refactoring to support
additional planes, and chroma planes with other-than 4:2:0
subsampling.
Change-Id: Idb76a0e23ab239180c818025bae1f36f1608bb23
Inside decode_sb_4x4 it should be
"get_tx_type_4x4(mb, y_idx * y_size + x_idx)"
but it was
"get_tx_type_4x4(mb, y_idx * (2 * y_size) + x_idx)".
Also making code of decode_sb_4x4, decode_sb_8x8, and decode_sb_16x16
formatted in the same way.
Change-Id: I15c7bef4fb575f7e9da19f953912324cb35d24dd
This function expects real Q values as inputs
not index values.
The use-age her impacts the Q chosen for force key
frames. Though this is a bug fix I have not yet verified
whether following the bug fix the q multiplier value used is
correct.
Change-Id: I49f6da894d90baeb1e86c820c335f02dc80d3b66
This code was only called in the BPRED case, but had no real special
case associated with it. Made BPRED behave like all other modes. No
bitstream change.
Change-Id: I87ba11fe723928b6314d094979011228d5ba006f
Took vp9_setup_scale_factors_for_frame() out from
vp9_setup_interp_filters(), so that it is only called once per
frame instead of per macroblock. Decoder tests showed a 1.5%
performance gain.
Change-Id: I770cb09eb2140ab85132f82aed388ac0bdd3a0aa
General code cleanup in loopfilter code. Modification of setup_frame_size,
so now VP9_COMMON is modified in one place after all width/height checks
passed.
Change-Id: Iedf32df43a912d7aae788ed276ac6c429973f6fe
Rename the file and clean up includes. In the future we would like to
pattern match the files which need additional compiler flags.
Change-Id: I2c76256467f392a78dd4ccc71e6e0a580e158e56
Adding decode_sb_8x8 and decode_sb_4x4 with common code for superblock
decoding. Renaming decode_superblock32 to decode_sb32 and
decode_superblock64 to decode_sb64.
Change-Id: Id006d7e398b9bfa3acec4326e1e0c537ebfefdd3
Using clamp and MIN/MAX functions instead of plain C code. Lower case
variable names. Removing redundant parenthesis.
Change-Id: Ibf7cc5fbe4fbdb5029049a599af71534176e6f42
We used to calculate SSIM only over the postproc buffer, whereas we
calculate PSNR for both. Compared to postproc-SSIM, this is about 0.3%
higher for derf, 1.4% lower for hd and 0.5% lower for stdhd, although
it is highly variable on a per-clip basis.
Change-Id: I8dd491f0f5b4201dedfb15d288c854d5d4caa10f
The patch adds the flexibility to use standard EOB based coding
on smaller block sizes and nzc based coding on larger blocksizes.
The tx-sizes that use nzc based coding and those that use EOB based
coding are controlled by a function get_nzc_used().
By default, this function uses nzc based coding for 16x16 and 32x32
transform blocks, which seem to bridge the performance gap
substantially.
All sets are now lower by 0.5% to 0.7%, as opposed to ~1.8% before.
Change-Id: I06abed3df57b52d241ea1f51b0d571c71e38fd0b
This threshold effectively limits the amount of motion
from one end of a GF/ARF group to the other.
This patch makes the threshold depend on image size.
Change-Id: Id45d1d7bced815f86ddd037be53164894b00b82f
Almost all arguments for vp9_build_inter32x32_predictors_sb and
vp9_build_inter64x64_predictors_sb can be deduced from the first macroblock
argument.
Change-Id: I5d477a607586d05698d5b3b9b9bc03891dd3fe83
Extracting setup_frame_size and update_frame_context functions. Introducing
vp9_read_prob function as shortcut for (vp9_prob)vp9_read_literal(r, 8).
Change-Id: Ia5c68fd725b2d1b9c5eb20f69cacb62361b5a3dd
Wrote sse2 version of vp9_short_idct_32x32 function. Compared
to c version, the sse2 version is 5X faster.
Change-Id: I071ab7378358346ab4d9c6e2980f713c3c209864
This gains about 0.2% on derf, 0.1% on hd and 0.4% on stdhd. I can put
this under an experimental flag if wanted, just trying to get my patch
queue in shape.
Change-Id: Ibe1a30fe0e0b07bec4802e0f3ff0ba22e505f576
Adds an experiment to use a weighted prediction of two INTER
predictors, where the weight is one of (1/4, 3/4), (3/8, 5/8),
(1/2, 1/2), (5/8, 3/8) or (3/4, 1/4), and is chosen implicitly
based on consistency of the predictors to the already
reconstructed pixels to the top and left of the current macroblock
or superblock.
Currently the weighting is not applied to SPLITMV modes, which
default to the usual (1/2, 1/2) weighting. However the code is in
place controlled by a macro. The same weighting is used for Y and
UV components, where the weight is derived from analyzing the Y
component only.
Results (over compound inter-intra experiment)
derf: +0.18%
yt: +0.34%
hd: +0.49%
stdhd: +0.23%
The experiment suggests bigger benefit for explicitly signaled weights.
Change-Id: I5438539ff4485c5752874cd1eb078ff14bf5235a
These are mostly just for experimental purposes. I saw small gains (in
the 0.1% range) when playing with this on derf.
Change-Id: Ib21eed477bbb46bddcd73b21c5c708a5b46abedc
Now that the first AC coefficient in both directions use the same DC
as their context, there no longer is a purpose in letting both have
their own band. Merging these two bands allows us to split bands for
some of the very high-frequency AC bands.
In addition, I'm redoing the banding for the 1D-ADST col/row scans. I
don't think the old banding made any sense at all (it merged the last
coefficient of the first row/col in the same band as the first two of
the second row/col), which was clearly an oversight from the band being
applied in scan-order (rather than in their actual position). Now,
coefficients at the same position will be in the same band, regardless
what scan order is used. I think this makes most sense for the purpose
of banding, which is basically "predict energy for this coefficient
depending on the energy of context coefficients" (i.e. pt).
After full re-training, together with previous patch, derf gains about
1.2-1.3%, and hd/stdhd gain about 0.9-1.0%.
Change-Id: I7a0cc12ba724e88b278034113cb4adaaebf87e0c
Pearson correlation for above or left is significantly higher than for
previous-in-scan-order (absolute values depend on position in scan, but
in general, we gain about 0.1-0.2 by using either above or left; using
both basically just makes this even better). For eob branch skipping,
we continue to use the previous token in scan order.
This helps about 0.9% on derf after re-training on a limited data set.
Full re-training and results on larger-resolution clips are pending.
Note that this commit breaks trellis, so we can probably get further
gains out of it by fixing trellis at some later point.
Change-Id: Iead68e296fc3a105cca746b5e3da9555d6010cfe
Reoptimizes the 8-tap smooth filter.
Results:
derf: +0.101%
yt: +0.157%
hd: +0.791%
stdhd: +0.264%
The next step will be to reoptimize the other two filters.
Change-Id: I3d256a510ad9c7c30c33fae4a70fb43dfc708ed0
Lower case variable names, declaration and initialization on the same line,
removing redundant casts to double.
Change-Id: I7ea3905bed827aa6faac11a78401b85e448b57f9
Lower case variable names, removing redundant variables, declaration and
initialization on the same line.
Change-Id: Ie0c6c95b14103990eb6a9d7784f8259c662e1251
Moving code from vp9_decode_frame function into setup_loopfilter and
setup_segmentation functions. A little bit of cleanup.
Change-Id: I2cce1813e4d7aeec701ccf752bf57e3bdd41b51c
Adds a per-frame, strength adjustable, in loop deringing filter. Uses
the existing vp9_post_proc_down_and_across 5 tap thresholded blur
code, with a brute force search for the threshold.
Results almost strictly positive on the YT HD set, either having no
effect or helping PSNR in the range of 1-3% (overall average 0.8%).
Results more mixed for the CIF set, (-0.5 min, 1.4 max, 0.1 avg).
This has an almost strictly negative impact to SSIM, so examining a
different filter or a more balanced search heuristic is in order.
Other test set results pending.
Change-Id: I5ca6ee8fe292dfa3f2eab7f65332423fa1710b58
Replaces the default tables for single coefficient magnitudes with
those obtained from an appropriate distribution. The EOB node
is left unchanged. The model is represeted as a 256-size codebook
where the index corresponds to the probability of the Zero or the
One node. Two variations are implemented corresponding to whether
the Zero node or the One-node is used as the peg. The main advantage
is that the default prob tables will become considerably smaller and
manageable. Besides there is substantially less risk of over-fitting
for a training set.
Various distributions are tried and the one that gives the best
results is the family of Generalized Gaussian distributions with
shape parameter 0.75. The results are within about 0.2% of fully
trained tables for the Zero peg variant, and within 0.1% of the
One peg variant.
The forward updates are optionally (controlled by a macro)
model-based, i.e. restricted to only convey probabilities from the
codebook. Backward updates can also be optionally (controlled by
another macro) model-based, but is turned off by default. Currently
model-based forward updates work about the same as unconstrained
updates, but there is a drop in performance with backward-updates
being model based.
The model based approach also allows the probabilities for the key
frames to be adjusted from the defaults based on the base_qindex of
the frame. Currently the adjustment function is a placeholder that
adjusts the prob of EOB and Zero node from the nominal one at higher
quality (lower qindex) or lower quality (higher qindex) ends of the
range. The rest of the probabilities are then derived based on the
model from the adjusted prob of zero.
Change-Id: Iae050f3cbcc6d8b3f204e8dc395ae47b3b2192c9
Lower case variable names, code simplification by using already defined
clamp and read_le16 functions.
Change-Id: I8fd544365bd8d1daed86d7b2ae0843e4ef80df08
As things stand the zero bin mode boost is hurting somewhat.
In part this seems to be because the boost applied as is
interferes with the rd mode selection loop.
Average gains (derf 0.072, yt 0.243, ythd 0.179 std-hd 0.212%)
Change-Id: Icaecea3908d9a7352370e49b8fa822f2c2c49dc1
Wrote sse2 version of vp9_short_idct10_16x16 function. Compared
to c version, the sse2 version is 2.3X faster.
Change-Id: I314c4f09369648721798321eeed6f58e38857f26
Making consistent initialization of mb_to_{top,botton,left,right}_edge
variables after set_mb_row & set_mb_col calls. A little bit of code cleanup
additionally.
Change-Id: I245bfe32c5701e9836956dc25cf8c770d109cbc1
Wrote sse2 version of vp9_short_idct16x16 function. Compared to c
version, the sse2 version is over 2.5X faster.
Change-Id: I38536e2b846427a2cc5c5423aaf305fd0e605d61
Renaming Width to width, Height to height and Version to version in
several structs and function signatures.
Change-Id: I084c3f7e747cb2ce3345aff27a3dff9b13a87543
Wrote sse2 functions of vp9_short_idct8x8 and vp9_short_idct10_8x8.
Compared to c version, the sse2 version is 2X faster. The decoder
test didn't show noticeable gain since 8x8 idct doesn't take much
of decoding time (less than 1% in my test).
Change-Id: I56313e18cd481700b3b52c4eda5ca204ca6365f3
If the intended display size is different than the size the frame is
coded at, then send that size explicitly in the bitstream. Adds a new
bit to the frame header to indicate whether the extra size fields
are present.
Change-Id: I525c66f22d207efaf1e5f903c6a2a91b80245854
Adjust the filter length and strength for each
ARF group based on a measure of difficulty (the boost)
and the active q range.
Remove lower limit on RDMULT value.
Average gains on the different sets in range 0.4%-0.9%.
However the ARNR changes give a very big boost on a
few clips.
Eg. Soccer ~5%, in derf set and Cyclist ~ 10% in the std-hd set
Change-Id: I2078d78798e27ad2bcc2b32d703ea37b67412ec4
This fix resolves some of the mismatch issues being seen
recently. While this is the right thing to do when tiling
is used for this experiment, it is not the underlying cause
of the the mismatches.
Something else is causing writing outside of the allowable
frame area in the encoder leading to this mismatch.
Change-Id: If52c6f67555aa18ab8762865384e323b47237277
These variables are unused, and are subject to overflowing, causing
assertions when built with -ftrapv.
Change-Id: Ia00a3201af309906c05bcd4b23a643925ed6ea86
If the second reference is better than the first in the long term,
it was possible to try to take the fractional exponent of a
negative number, giving an undefined result.
Change-Id: I1dd08286747ceae960eb03bb5d98a383cc9d253b
Moving identical code to separate functions, variable declaration and
initialization on the same line.
Change-Id: Ifa6474a64189f9d8051e88e19850453b0227752c
Updates the YV12_BUFFER_CONFIG structure to be crop-aware. The
exiting width/height parameters are left unchanged, storing the
width and height algined to a 16 byte boundary. The cropped
dimensions are added as new fields.
This fixes a nasty visual pulse when switching between scaled and
unscaled frame dimensions due to a mismatch between the scaling
ratio and the 16-byte aligned sizes.
Change-Id: Id4a3f6aea6b9b9ae38bdfa1b87b7eb2cfcdd57b6
This is like VP8_COPY_REFERENCE, but returns a pointer to the reference
frame rather than a copy of it. This is useful when the application
doesn't know what the size of the reference is, as is the case when
scaling is in effect.
Change-Id: I63667109f65510364d0e397ebe56217140772085
Enable entropy coding of motion vectors up to +/-2048. Also
extend the motion search range accordingly.
Change-Id: Iac2bb015e8934521cef83a19edbe967d9f097436
The superframe index marker byte carries data in the lower 5 bits. Only the
upper 3 should be used as part of the mask to detect it. By masking with
0xf0, the previous code was incorrect for frames over 65k bytes.
Change-Id: I6248889f5af227457f359a56b2348ef6db87a3b4
If a superframe (ARF) is generated while flushing the lagged frames
at the end of the clip, the buffer pointer wasn't being properly
updated to account for the size of the index, causing the next
frame to overwrite the index on the previous frame.
Change-Id: Ib158cc8e4183d663bdfb9ba002dd4c98916abdc9
If the bool-coded partition naturally ends in a byte that matches the
superframe index marker, it could lead to a parse error. This commit
ensures that if such a marker is seen, it is padded out with an
additional zero byte to disambiguate it.
Change-Id: Id977de05745b6fa9ef08afb71e210a2a3ecca02e
When coding the frame that corresponds to the midpoint frame
defining an ARF, do not update the last reference frame buffer.
Previously this buffer was updated meaning that when coding the next
ARF all the reference buffers were the same (or nearly so).
Turning the update off means that the frame before is still available
as an alternative predictor and for use in compound prediction.
Also fixed inconsistency in test for mismatch (patch from JK).
Net average gains (derf 0.049, yt 0.163, yt-hd 0.207, std-hd 0.286)
Change-Id: Ifee21da21ccbb1648ac2eafe890d3ce60562c7bc
Removing redundant code, introducing new functions for better
decomposition, adding 'clamp' function to vp9_common.h.
Change-Id: Ic3b8ca13bbc38f60f0c9c43910b5802005e31aaf
This patch puts in an adjustment to the maximum gf/arf
interval based on the active q range. It sets a fixed
baseline maximum of 16 but can drop this down to 12 at
lower q. This required some re-ordering in the first pass
code to insure we have a Q range estimate before defining
the first gf sequence.
The main gains seed are int he STD hd set on 50fps clips
where previously the interval could rise as high as 25.
On the std hd clip the gains are around 2.8% with limit set
to 300 frames.
When combined with the one shot rate control flags we get
combined of:
derf 1.55% (limit300), yt 7.25%, hd 5.17% std-hd 5.84% (limit300)
Change-Id: Ib380d51354511f2ff0f171a8df4e74291c0421f9
Remove the temporary branch count arrays and build the adapted probabilities
while walking the tree. Gives an additional 1.5% or so on CIF.
Change-Id: I875d61e5e0ec778e5d2f7f9d0837b989a91cf3a3
The previous implementation visited each node in the tree multiple times
because it used each symbol's encoding to revisit the branches taken and
increment its count. Instead, we can traverse the tree depth first and
calculate the probabilities and branch counts as we walk back up. The
complexity goes from somewhere between O(nlogn) and O(n^2) (depending on
how balanced the tree is) to O(n).
Only tested one clip (256kbps, CIF), saw 13% decoding perf improvement.
Note that this optimization should port trivially to VP8 as well. In VP8,
the decoder doesn't use this function, but it does routinely show up
on the profile for realtime encoding.
Change-Id: I4f2848e4f41dc9a7694f73f3e75034bce08d1b12
Adds probability updates for extra bits for the nzcs, code for
getting nzc stats, plus some minor cleanups and fixes.
Change-Id: If2814e7f04fb52f5025ad9f400f3e6c50a00b543
