summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2014-09-22Remove unnecessary local variable declarationJingning Han
This commit removes a repetitive local variable declaration in vp9_rd_pick_inter_mode_sb. Change-Id: I1b0afa98ff1ecbfb46e17d3d1cee95d32c4309db
2014-09-22Adaptive mode search schedulingJingning Han
This commit enables an adaptive mode search order scheduling scheme in the rate-distortion optimization. It changes the compression performance by -0.433% and -0.420% for derf and stdhd respectively. It provides speed improvement for speed 3: bus CIF 1000 kbps 24590 b/f, 35.513 dB, 7864 ms -> 24696 b/f, 35.491 dB, 7408 ms (6% speed-up) stockholm 720p 1000 kbps 8983 b/f, 35.078 dB, 65698 ms -> 8962 b/f, 35.054 dB, 60298 ms (8%) old_town_cross 720p 1000 kbps 11804 b/f, 35.666 dB, 62492 ms -> 11778 b/f, 35.609 dB, 56040 ms (10%) blue_sky 1080p 1500 kbps 57173 b/f, 36.179 dB, 77879 ms -> 57199 b/f, 36.131 dB, 69821 ms (10%) pedestrian_area 1080p 2000 kbps 74241 b/f, 41.105 dB, 144031 ms -> 74271 b/f, 41.091 dB, 133614 ms (8%) Change-Id: Iaad28cbc99399030fc5f9951eb5aa7fa633f320e
2014-09-19Merge "Remove mi_grid_* structures."hkuang
2014-09-19Remove mi_grid_* structures.hkuang
mi_grid_* are arrays of pointer to pointer. They save the pointers that point to the MIs in cm->mi. But they are unnecessary and complicated. The original goal was to remove MODE_INFO_t copy. But with an extra MODE_INFO_t pointer inside MODE_INFO_t, same goal could be achieved. This commit totally removes the mi_grid_* structures. But there are still many dummy MODE_INFO_t inside cm->mi which are a waste of memory. Next commit will do on-demand MODE_INFO_t allocation in order to save these memories. Change-Id: I3a05cf1610679fed26e0b2eadd315a9ae91afdd6
2014-09-19Merge "test/tools_common.sh: Add support for dumping list of tests."James Zern
2014-09-19Merge "High bit-depth coefficient coding functions"Deb Mukherjee
2014-09-19test/tools_common.sh: Add support for dumping list of tests.Tom Finegan
Via new command line argument --list-tests. Change-Id: I0f12d10b49c14f5f68aafd5bc45362b1ea4f361f
2014-09-19High bit-depth coefficient coding functionsDeb Mukherjee
Tokenization and Detokenization enhancements for 10/12 bit Change-Id: I3c269ec30f8eb160ee024905638a193975237559
2014-09-19Merge "build/make/iosbuild.sh: Support extra configure args."Tom Finegan
2014-09-19Merge "build/make/rtcd.pl: Add missing arm64 support."Tom Finegan
2014-09-19build/make/iosbuild.sh: Support extra configure args.Tom Finegan
Change-Id: I9860ac1bb0180315e26987c5fc53d814ea1f0283
2014-09-19Merge "rtcd: relocate 'extern "C" {' declaration"James Zern
2014-09-19build/make/rtcd.pl: Add missing arm64 support.Tom Finegan
Treats arm64 the same as armv8. Change-Id: Ia5029f7bbff251d5aa0836bcccdf20fa0a693891
2014-09-19Merge "vp8: Update to the adaptive denoising mode."Marco
2014-09-19vp8: Update to the adaptive denoising mode.Marco
Allow for option to apply spatial blur for temporal denoising, under the aggressive denoising mode. Change-Id: I41c5fdc0b6cf32d8f8d1d4236b25fa5aa406e89e
2014-09-18rtcd: relocate 'extern "C" {' declarationJames Zern
Place after pulling in forward declarations from the codec. This fixes compilation of the tests under vs9. Since 10783d4 Adds high bitdepth transform functions and tests where vp9_idct.h was added to vp9_rtcd.h the tests are pulling in vp9_systemdependent.h, which under visual studio include intrin.h. With VS9 these include headers which define helper classes for intel intrinsics. When including it in the tests (via vp9_rtcd.h) __cplusplus is defined but vp9_rtcd.h would wrap declarations in 'extern "C" {' causing a mismatch in linkage which resulted in compilation failure. Change-Id: I475e50198b71320e8606bc95c9454876d8799ede
2014-09-18Merge "Add a conditional compilation macro to VP8 unit test."James Zern
2014-09-18Add a conditional compilation macro to VP8 unit test.JackyChen
In many tests in VP8, the denoiser is disabled. By adding this conditional comilation macro, the unit test will not be included when denoiser is not enabled. Change-Id: I6edec85c996acca22aacd11161c52408be2660a3
2014-09-18Merge "FIX: vp9_loopfilter_intrin_sse2.c"Frank Galligan
2014-09-18FIX: vp9_loopfilter_intrin_sse2.cScott LaVarnway
Fixes Visual Studio build failures Change-Id: I233719cd63b3ad0db16e2834bf1d7ea1df805880
2014-09-18Merge "Add unit test for VP9 denoiser."JackyChen
2014-09-18Merge "[spatial svc] Use same golden frame for all temporal layers"Minghai Shang
2014-09-18Add unit test for VP9 denoiser.JackyChen
Check basic datarate targeting for a single bitrate when the denoiser is on. Change-Id: Iad0e664859ffeb4ae6dd5f6dfb71a121e917e3fb
2014-09-18[spatial svc] Use same golden frame for all temporal layersMinghai Shang
Overhead goes down from 8% to 3% for 1080 60p Change-Id: Idf3e5ca8712402a914a8cb79df17d3cdab63b163
2014-09-18Merge "Adds high bitdepth convolve, interpred & scaling"Deb Mukherjee
2014-09-18Adds high bitdepth convolve, interpred & scalingDeb Mukherjee
Change-Id: Ie51c352a6b250547207cbc1ebba833a01ed053e3
2014-09-18Substantial reworking of code for arf and kf groups.Paul Wilkins
Substantial restructuring of the way we estimate the rate of decay in prediction quality and determine the arf interval and amount of boost used. Also other changes to support moving to a lower first pass Q which exposes some new features and allows us to better distinguish genuinely static blocks from low motion or noisy blocks. Net gains now visible on all the test sets with std-hd PSNR up 1.87%. There are still some bad outlier cases but most of these are low motion or slide show type content where the metrics are already high at any given rate. The best + case is up by more than 10%. Change-Id: I18e25170053bdf3188f493ff8062f48a74515815
2014-09-17Merge "Improved mb_lpf_horizontal_edge_w_sse2_16() #2"Frank Galligan
2014-09-17Improved mb_lpf_horizontal_edge_w_sse2_16() #2Scott LaVarnway
The decoder performance improved up to 1% for the test clips used. Change-Id: I4621112bdccfba01640322facfa4ba8da8290ea5
2014-09-17Merge "Resolves a few gcc warnings"Deb Mukherjee
2014-09-17Merge "vp8: common: postproc: fix signed overflow of statement of (X +c) >= ↵Johann
X when '-Werror=strict-overflow' is set."
2014-09-16Resolves a few gcc warningsDeb Mukherjee
clang is fine. Change-Id: Ia4e9ff17ea3b86bc87dca35828ee7ce45bea6994
2014-09-16Merge "arm: Fix building vp8_mse16x16_neon.c with MSVC"James Zern
2014-09-16Merge "Adding high-bitdepth intra prediction functions"Deb Mukherjee
2014-09-16Merge "Remove memset of every external frame buffer."Frank Galligan
2014-09-16Merge "Speeding up decode_coeffs()."Dmitry Kovalev
2014-09-16Adding high-bitdepth intra prediction functionsDeb Mukherjee
Change-Id: I6f5cb101e2dc57c3d3f4d7e0ffb4ddbed027d111
2014-09-16Adds high bitdepth quantization functionsDeb Mukherjee
Adds various high bitdepth quantization functions. Change-Id: I36fc0bf75a1bd15128ed271df8723de0ac134b0c
2014-09-16Merge "Use non-zero mode threshold for NEARESTMV modes"Jingning Han
2014-09-16Merge "Fix a performance regression"Yaowu Xu
2014-09-16arm: Fix building vp8_mse16x16_neon.c with MSVCScott LaVarnway
Use the right return values - vadd_s64 returns int64x1_t, not a normal int64_t. Change-Id: Ife17213087c1dfb5faaa647f804d2fd140f3a0eb
2014-09-16Merge "Fix ARF construction when scaling"Adrian Grange
2014-09-16Speeding up decode_coeffs().Dmitry Kovalev
1080p decoding speedup -- 1.25%, 4K decoding speedup -- 2.5%. Change-Id: I5f02f521cbf7758d4d1886a93bc5b074abdff03d
2014-09-16Merge "Move call to vp9_rc_get_second_pass_params()."Adrian Grange
2014-09-16Fix a performance regressionYaowu Xu
This commit adds back sse2 or ssse3 optimized versio of a couple of functions, fixes a ~10% performance regression. Change-Id: I049786906e5a641224dced63c6492aec9d86d183
2014-09-16Fix ARF construction when scalingAdrian Grange
The ARF frame should always be the same size as the native resolution of the input frames. It will be scaled to the required resolution at encode time. Change-Id: I0afe858129aa6ef65b1648f43476331715346896
2014-09-16Merge "build/make/iosbuild.sh: Be more explicit about armv7."Tom Finegan
2014-09-16Use non-zero mode threshold for NEARESTMV modesJingning Han
This commit makes the encoder to use non-zero mode threshold for NEARESTMV modes. The runtime for test clips of speed 3 is reduced by about 1%. pedestrian 1080p 2000 kbps, 143239 ms -> 141989 ms bus CIF 1000 kbps, 7835 ms -> 7749 ms The compression performance change is about -0.02% for both derf and stdhd. Change-Id: Ib71808922c41ae2997100cb7c561f68dcebfa08e
2014-09-15Merge "Add ARF validation for compound inter mode check"Jingning Han
2014-09-15Merge "Remove redundant reference frame check in sub8x8 RD search"Jingning Han
generated by cgit v1.2.3 (git 2.25.1) at 2024-10-10 21:31:42 +0000