generic-library/vpx
1
0
Fork 0
Code Issues Pull Requests Releases Wiki Activity
732 Commits 55 Branches 16 Tags 63 MiB
eed2ce58e3
Commit Graph

567 Commits

Author SHA1 Message Date
John Koleszar
eed2ce58e3 Merge "Fix counter of fixed keyframe distance" 2011-03-04 05:28:38 -08:00
Johann
6f5189c044 Merge "ARMv6 optimized half pixel variance calculations" 2011-03-02 05:48:46 -08:00
Yunqing Wang
cfaee9f7c6 Merge "Add prefetch before variance calculation" 2011-02-28 11:42:28 -08:00
Scott LaVarnway
3e6d476ac3 Merge "Avoid double copying of key frames into alt and golden buffer" 2011-02-28 10:16:33 -08:00
Yunqing Wang
d96ba65a23 Add prefetch before variance calculation 
This improved encoding performance by 0.5% (good, speed 1) to
1.5% (good, speed 5).

Change-Id: I843d72a0d68a90b5f694adf770943e4a4618f50e
 2011-02-28 11:25:55 -05:00
Johann
31dab574cc Merge "Remove a second check for invalid ptr in vp8_get_compressed_data" 2011-02-25 11:44:18 -08:00
Johann
e4fa638653 Merge "Remove temporal alt ref from realtime only build" 2011-02-25 06:55:17 -08:00
Johann
1fae7018a8 Merge "Handle mem allocation failure in vp8e_init" 2011-02-25 06:55:10 -08:00
Attila Nagy
d8fc974ac0 Avoid double copying of key frames into alt and golden buffer 
Change-Id: I726976a297a593a35ed6cba3c660e372562f7b27
 2011-02-25 09:03:16 +02:00
Attila Nagy
6da2018789 Remove a second check for invalid ptr in vp8_get_compressed_data 
Check is done first when function si entered.

Change-Id: Ief0d0cbd4860aaf492b78728f8d22f24029b1174
 2011-02-25 08:41:13 +02:00
Scott LaVarnway
861175ef00 Removed vp8_block2type 
and used defines instead.

Change-Id: Idb56e0295d004793f406dfd2d8d8c546aad62e03
 2011-02-24 14:35:18 -05:00
Scott LaVarnway
d53492bba4 Merge "Revisited rd_pick_intra4x4block" 2011-02-24 11:25:21 -08:00
Scott LaVarnway
658454a04c Revisited rd_pick_intra4x4block 
Removed unnecessary copies.  No noticeable speed gains.


Change-Id: I996c50c23fedd06d54ee7a3e762cbf559cc4a9d1
 2011-02-24 13:31:47 -05:00
Paul Wilkins
b862c108dd Overflow of frame error accumulators. 
This fixes an overflow problem in the frame error accumulators.

The overflow condition is extreme but did trigger when Frank B.
coded some high motion interlaced HD content.

The observed effect was a catastrophic  breakdown of the rate
control leading to massive undershoot and poor bit allocation.

All the error values should really be unsigned but I will look at this
separately.

Change-Id: I9745f5c5ca2783620426b66b568b2088b579151f
 2011-02-24 15:49:41 +00:00
Tero Rintaluoma
8ae92aef66 ARMv6 optimized half pixel variance calculations 
Adds following ARMv6 optimized functions to the encoder:
 - vp8_variance_halfpixvar16x16_h_armv6
 - vp8_variance_halfpixvar16x16_v_armv6
 - vp8_variance_halfpixvar16x16_hv_armv6

Change-Id: I1e9c2af7acd2a51b72b3845beecd990db4bebd29
 2011-02-23 13:27:27 +02:00
Attila Nagy
e6db21ecc4 Handle mem allocation failure in vp8e_init 
Change-Id: I0d0445c57eb0889082f83de1948852d57b38fefb
 2011-02-23 12:36:03 +02:00
Attila Nagy
7af0d906e3 Remove temporal alt ref from realtime only build 
It is not used in realtime mode. Reduces memory footprint.

Change-Id: I7f163225762368df5457cfd413050161d3704a3f
 2011-02-22 12:53:32 +02:00
Johann
945dad277d Revert "use unaligned load" 
This reverts commit f50f2fd2a7.

Change Ib7506e3e aligns the buffer

Change-Id: Ie0f8bd3e57cfdfef81d39638a1451458ebbae2e0
 2011-02-18 10:23:02 -05:00
John Koleszar
c764c2a20f Merge "clean up unused files" 2011-02-18 06:33:05 -08:00
John Koleszar
3ed8fe8778 remove unused vp8_predict_dc function 
Change-Id: I64fa47889c54cfed094a674c49ef0996d49bdd42
 2011-02-18 09:12:20 -05:00
John Koleszar
cbf923b12c clean up unused files 
Removed a number of files that were unused or little-used.

Change-Id: If9ae5e5b11390077581a9a879e8a0defe709f5da
 2011-02-18 09:09:49 -05:00
John Koleszar
d371ca93e5 cosmetic: remove unnecessary scope 
Clean up some unnecessary scoping around pick_filter_level.

Change-Id: Ic57fa33e3fcae37fe6beae977e5743783399d5af
 2011-02-18 08:46:07 -05:00
John Koleszar
597d02b508 Merge "Dont pick encoder filter level when loopfilter is disabled." 2011-02-18 05:26:23 -08:00
Attila Nagy
fb5a692d27 Reinitialize quantizer only when any delta is changing 
No need to reinitialize for base Q changes.

Change-Id: Ie76ec21dd3c5582d5183dbed75ed73a1eed3e291
 2011-02-18 14:23:37 +02:00
Attila Nagy
c6ef75690f Dont pick encoder filter level when loopfilter is disabled. 
Change-Id: I58154faf4f3ece24f9927a5c3ab7e830e0887fb6
 2011-02-18 08:53:00 +02:00
John Koleszar
b2ae57f1b6 Merge "Use endian-neutral bitstream packing/unpacking" 2011-02-17 12:34:16 -08:00
John Koleszar
562f1470ce Use endian-neutral bitstream packing/unpacking 
Eliminate unnecessary checks on target endianness and associated
macros.

Change-Id: I1d4e6a9dcee9bfc8940c8196838d31ed31b0e4aa
 2011-02-17 15:20:53 -05:00
John Koleszar
ac10665ad8 Merge "Removed unused vp8_recon_intra4x4mb function" 2011-02-17 11:30:13 -08:00
Scott LaVarnway
07f7b66fae Removed unused vp8_recon_intra4x4mb function 
Change-Id: I4a328ce152d9dbe6b0d1606d1b523e8e7bfb468e
 2011-02-17 13:34:38 -05:00
John Koleszar
c351aa7f1b Merge "Fix relative include paths" 2011-02-17 04:13:44 -08:00
Yunqing Wang
da9402fbf6 Merge "Allocate source buffers to be multiples of 16" 2011-02-16 11:35:06 -08:00
Yunqing Wang
da227b901d Allocate source buffers to be multiples of 16 
Currently, when the video frame width is not multiples of 16, the
source buffer has a stride of non-multiples of 16, which forces
an unaligned load in SAD function and hurts the performance. To
avoid that, this change allocates source buffers to be multiples
of 16.

Change-Id: Ib7506e3eb2cea06657d56be5a899f38dfe3eeb39
 2011-02-16 12:57:17 -05:00
Johann
0c2cfff9b0 Merge "ARMv6 optimized sad16x16" 2011-02-16 05:22:38 -08:00
James Zern
0030303b69 Remove redundant ptr checks in calls to vpx_free 
vpx_free if used contains this check. If replaced, well behaved free
will behave similarly.

Change-Id: I25483aaa8b39255b9a8cf388d6e5eaa20a908ae1
 2011-02-15 12:43:35 -08:00
Yunqing Wang
7725a7eb56 Merge "Improve vp8_sad16x16_sse3 function" 2011-02-14 14:09:25 -08:00
Yaowu Xu
27dad21548 Merge "Improved vp8_rd_pick_intra_mbuv_mode" 2011-02-14 13:58:12 -08:00
Scott LaVarnway
94d4fee08f Improved vp8_rd_pick_intra_mbuv_mode 
Eliminated unnecessary calculations. Very small change
to performance.

Change-Id: Ib7213d43c64e36955177c4d47950ff472266f822
 2011-02-14 16:34:33 -05:00
Yunqing Wang
2debd5b5f7 Improve vp8_sad16x16_sse3 function 
In real-time mode, vp8_sad16x16 function is called heavily in
motion search part. Improvement of this function gives 1.2%
encoding performance gain (real-time mode, tulip clip).

Change-Id: I23c401fc40c061f732a9767e8d383737a179bd58
 2011-02-14 16:23:49 -05:00
Yaowu Xu
404e998eb7 Merge "mem leak fix for cpi->tplist" 2011-02-14 11:29:22 -08:00
James Berry
d3dfcde0f7 mem leak fix for cpi->tplist 
checks added to make sure that cpi->tplist
is freed correctly in vp8_dealloc_compressor_data
and vp8_alloc_compressor_data.

Change-Id: I66149dbbd25c958800ad94f4379d723191d9680d
 2011-02-14 14:02:52 -05:00
Scott LaVarnway
d419b93e3e Improved rd_pick_intra4x4block 
Eliminated unnecessary calculations.  Improved performance
by 10% on keyframes and 1.6% overall for the test clip used.

Change-Id: I87671b26af5e2cc439e81d0fee3b15c7cd2a3309
 2011-02-14 13:32:58 -05:00
Johann
0ff10bb1f7 Merge "remove assembly detokenizer" 2011-02-14 05:10:16 -08:00
Johann
bb6bcbccda remove assembly detokenizer 
hasn't been kept up to date. remove it to avoid confusion.

Change-Id: I52ffde19b59fec5c7a381299ca2e85cb38330be7
 2011-02-11 11:09:00 -05:00
Yunqing Wang
353246bd60 Merge "Add improved_mv_pred flag in real-time mode" 2011-02-11 07:20:17 -08:00
Yunqing Wang
9d0b2cbbce Add improved_mv_pred flag in real-time mode 
As mentioned in check-in "Improve motion search in real-time mode",
MV prediction calculation causes speed loss for speed 7 and above.
This change added a flag to turn off this calculation for speed>6
in real-time mode.

Change-Id: I9f4ae5a8bf449222d1784b54e7d315fc8347b2d1
 2011-02-11 09:59:41 -05:00
Tero Rintaluoma
1ef86980b9 ARMv6 optimized sad16x16 
Adds a new ARMv6 optimized function vp8_sad16x16_armv6 to encoder.

Change-Id: Ibbd7edb8b25cb7a5b522d391b1e9a690fe150e57
 2011-02-11 11:14:07 +02:00
Yaowu Xu
4f8a166058 Merge "Redefining good quality speed settings" 2011-02-10 21:38:19 -08:00
Yunqing Wang
6f53e59641 Merge "Improve motion search in real-time mode" 2011-02-10 12:42:44 -08:00
John Koleszar
02321de0f2 Fix relative include paths 
Allow compiling without adding vp8/{common,encoder,decoder} to the
include paths.

Change-Id: Ifeb5dac351cdfadcd659736f5158b315a0030b6c
 2011-02-10 15:09:44 -05:00
Yunqing Wang
41e6eceb28 Improve motion search in real-time mode 
Applied better MV prediction in real-time mode, which improves
the encoding quality.

Used quarter-pixel search instead of iterative sub-pixel search
for speed >=5 to improve encoding performance.

Tests on the test set showed:
1. For speed=-5, quality improvement: 1.7% on AvgPSNR and 2.1%
on SSIM, performance improvement: 3.6% (This counts in the
performance lose caused by MV prediction calculation in "Improve
MV prediction in vp8_pick_inter_mode() for speed>3").
2. For speed=-8, quality improvement: 2.1% on AvgPSNR and 2.5%
on SSIM. but, 6.9% performance decrease because of MV prediction
calculation. This should be improved later.

Change-Id: I349a96c452bd691081d8c8e3e54419e7f477bebd
 2011-02-10 13:40:24 -05:00