Age | Commit message (Collapse) | Author |
|
|
|
|
|
Reset the last_coded_q_map and the sb->index in the cyclic_refresh
on a re-encode for slide change, so the refresh can start again
right after slide change.
Change-Id: I10cbc8354de8f7c2863b4212e6793b58a048b330
|
|
|
|
Add scene detection flag to choose_partitioning to force split
of 64x64 block partition. This reduces artifacts on slide changes.
Bug:b/110978869
Change-Id: I9cc79a7c03f3aa2edeb28656b09a2177b72d59a8
|
|
Add partition search breakout models for 720p resolution,
currently enabled only for speed 0.
Compression performance change is neutral.
Tested encoding speed over 20 720p clips:
Speed gain(%) QP=55 QP=45 QP=35
max 22.1 20.3 29.8
average 10.3 9.1 11.4
Change-Id: I07499728bbc5b80035fc66fad882ea556c8d07f2
|
|
Change-Id: I26ced25ff2e20ec414d5ecaa7d26f4a69175896c
|
|
|
|
|
|
|
|
|
|
Use the avg_frame_low_motion to reduce/turnoff this
early exit for higher motion content. Get some quality
back for higher motion clips and keep the same exit
thresh for low motion clips.
Change-Id: I95daf754dc0048b3e935d1a753f7f1101e6ffb77
|
|
Change-Id: I888761193882cc92720e0efaea5229a04a6ed67f
|
|
ARGBToRGB24Row_AVX512VBMI fails to compile on Mac:
row_gcc.cc: instruction requires: AVX-512 VBMI ISA AVX-512 VL ISA
BUG=libyuv:789
Change-Id: Ibd584e8c82e3ce86ec5460b4243f84f5dbdf4c81
|
|
If the rate cost returns as INT_MAX, directly set the rdcost as
INT64_MAX.
Change-Id: I3ea1963aff10040dd9cef805beed9aebeedb93bc
|
|
This patch relates to motion artifacts as described in Issue 73484098
The aim of this patch is to promote the use of smaller partition
sizes in places where some of the sub blocks have very low
spatial complexity and some have much higher complexity.
The patch can have a small impact on encode speed, but much
less than alternative approaches such as lowering the rd thresholds
that limit the partition search when distortion is low.
The patch also applies a similar sub block strategy for AQ1.
Metrics results for our standard sets over typical YT rates.
(Overall PSNR, SSIM, PSNR HVS) % -ve better.
Low Res -0.274, -0.303, -0.330
Mid Res 0.001, - 0.128, -0.100
Hd Res -0.236, -0.371, -0.349
N 2K -0.663, -0.798, -0.708
N 4K -0.488, -0.588, -0.517
Change-Id: Ice1fc977c1d29fd5e401f9c7c8e8ff7a5f410717
|
|
|
|
|
|
|
|
The avg_frame_low_motion metric is only computed on the
top spatial layer, and since its part of the layer context
struct, it needs to written to all lower spatial layers for
consistency.
Small/minor change in metrics.
Change-Id: I92a001c37aeb332e613212288b13a2ed9745af88
|
|
For SVC: apply the sse_zeromv early exit also to
the case where golden is second temporal reference.
Set the thresh_svc_golden threshold for this case.
This is reduce the encode time for case where golden
is second temporal reference for SVC.
Change-Id: I8c0c87dd746579d3c4f5e983c7f9dd0a1e1476e0
|
|
Speed tests are disabled by default.
Change-Id: I49f8da3d3e1e4d9c72b17fc47c098284e7d84236
|
|
Change-Id: I50b32f37f77224ebf0470545152c83ae2ed3cfa3
|
|
|
|
|
|
* changes:
[VSX] Replace vec_pack and vec_perm with single vec_perm
VSX Version of fdct32x32_rd
|
|
|
|
vpx_quantize_b:
VP9QuantizeTest Speed Test (POWER8 Model 2.1)
32x32 Old VSX time = 8.1 ms, new VSX time = 7.9 ms
vp9_quantize_fp:
VP9QuantizeTest Speed Test (POWER8 Model 2.1)
32x32 Old VSX time = 6.5 ms, new VSX time = 6.2 ms
Change-Id: Ic2183e8bd721bb69eaeb4865b542b656255a0870
|
|
Low bit depth version only. Passes the Trans32x32Test test suite.
Trans32x32Test Speed Test (POWER9 Model 2.2)
32x32 C time = 212.7 ms (±0.1 ms), VSX time = 82.3 ms (±0.0 ms) [2.6x]
Change-Id: If906ec9b56ce3818cae0cc462c7277284ab29859
|
|
|
|
|
|
bit_depth_ was not initialized (used in FillRandom)
and caused valgrind errors.
BUG=webm:1542
Change-Id: I09a9acd54de0dfa4f9006304f45eb20883c9908c
|
|
|
|
|
|
|
|
No need to pass arguments that are already members of the class.
Change-Id: I887d33d6037b561dee5dd8d49bb112d9120cd2a7
|
|
Change-Id: I4d8b6d114d9a407f5bb879ab059a66425976f1df
|
|
BUG=webm:1542
Change-Id: If3e0b32a6832740b9af2f5c2d9418a6664297f57
|
|
Enable ML based partition search breakout on speed 0 when frame
resolution is less then 720p and bitdepth is 8.
Compression performance change is neutral.
Tested encoding speed over 20 480p sequences:
Speed gain(%) QP=30 QP=40 QP=50 QP=60
max 14.4 18.6 17.8 24.4
average 4.6 9.0 8.0 13.2
Change-Id: Ia0d2947030ac774dc1533eb27ffc57f5b788a6ce
|
|
Disable the cyclic refresh for very low average Q.
This reduces encoded bitrate for static slides after the
the quality has ramped up well enough (low Q). And as the
cyclic refresh is not needed at low Q in most cases, this
has minimal/no effect on quality on RTC set.
Change-Id: Id6d449aa2351bb6886d72aafb2d406e967ed2789
|
|
|
|
|
|
|
|
|
|
Fixes to nonrd coding mode for lossless mode: keep
skip_txfm to 0 (no skip) and disable the encoder breakout.
This makes the encoding lossless when that mode is selected
for real-time (nonrd pickmode).
Also the disable the cyclic refresh for lossless mode.
Change-Id: I20a11ef6df08accec472d26fabebd14d51f4d337
|
|
Fix condition in frame dropper for SVC to handle case
where spatial layer is skipped encoded (due to 0 bitrate).
Change-Id: I24185178774d73e8bb1c406acc0292422dfbe174
|
|
for q-index between 100 and 150.
This only affects speed 1 and 2, resolution under 720p, q-index between
100 and 150, low bit-depth.
Compression performane change is neutral.
Encoding speed gain is up to 16% for speed 1;
up to 6% for speed 2.
Results from encoding city_4cif_30fps:
speed 1, QP=36
before: 37.964 dB, 45581b/f, 2.73 fps
after: 37.958 dB, 45510b/f, 3.16 fps
speed 1, QP=28
before: 39.297 dB, 82452b/f, 2.14 fps
after: 39.297 dB, 82310b/f, 2.25 fps
speed 2, QP=36
before: 37.903 dB, 45586b/f, 4.08 fps
after: 37.895 dB, 45492b/f, 4.34 fps
speed 2, QP=28
before: 39.224 dB, 82272b/f, 3.03 fps
after: 39.223 dB, 82152b/f, 3.17 fps
Change-Id: Ieaefedad902df80aa9699545fa06294601955803
|
|
|
|
Create tests for sync layer. The purpose of new tests is not to check
bitrate targeting, thus they're put in a new file.
Create a base class for svc tests, which is also inherited by svc datarate
tests, to reduce code redundancy.
Start decoding in the test from the frame of layer sync.
Change-Id: I7226d208279ad785873dffef51e0a8abef23b256
|
|
Previous CLs have implemented the construction of the hierarchical
structure at the encoder side. This CL is to define and configure the
according flags that will guide the reference frame update according to
the constructed hierarchical structure.
Change-Id: Iae55f2400f7c7beff41feff9308f87bfc70c7b21
|