Log of /xvidcore/src/motion/motion_est.c
Parent Directory
Revision
1.58.2.36
Wed Sep 10 22:19:00 2003 UTC (20 years, 7 months ago) by
edgomez
Branch:
dev-api-4
Changes since
1.58.2.35: +1 -1 lines
FILE REMOVED
Motion Estimation module splitting.
Revision
1.58.2.33 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Thu Sep 4 18:40:02 2003 UTC (20 years, 7 months ago) by
Isibaar
Branch:
dev-api-4
Changes since
1.58.2.32: +458 -6 lines
Diff to
previous 1.58.2.32
, to
branch point 1.58
Newly introduced:
- XVID_ME_DETECT_STATIC_MOTION: skips MV search in stationary scenes (gives a slight speed-up and a PSNR drop of ~0.03 dB)
- XVID_ME_FASTREFINE16: low complexity subpel refinement. Useful mainly for qpel (big speed-up, small PSNR drop of ~0.05 dB)
- XVID_VOP_FAST_MODEDECISION_RD: performs R-D checking only 'when really needed' (significant speed-up, again mild PSNR drop)
All new options need further testing/fine tuning...
Revision
1.58.2.10 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Thu Apr 10 13:05:54 2003 UTC (21 years ago) by
edgomez
Branch:
dev-api-4
Changes since
1.58.2.9: +8 -8 lines
Diff to
previous 1.58.2.9
, to
branch point 1.58
2003-04-10 GMT Edouard Gomez <ed.gomez@free.fr> patch-14
Summary:
Removed all ABS() macros.
Revision:
xvidcore--devapi4--1.0--patch-14
All ABS macros have been replace with their stdlib.h/math.h
equivalent. This gives a 33% overall speedup for the plain C encoder,
while the ia32 one seems to suffer a small speed loss. However this
speed loss is very small and it seems it depends on the CPU type as
the abs/fabs usage is impacting badly on sad functions but impacting
well on interpolate functions ... weird inst'it ?
modified files:
src/bitstream/mbcoding.c src/global.h src/image/image.c
src/motion/motion_comp.c src/motion/motion_est.c
src/motion/sad.c src/prediction/mbprediction.c
src/utils/mbtransquant.c src/xvid.c
Revision
1.50 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Sat Sep 21 11:59:22 2002 UTC (21 years, 7 months ago) by
edgomez
Branch:
MAIN
Changes since
1.49: +11 -4 lines
Diff to
previous 1.49
- Disabled bframe specific code and macros. My tests seem to confirm this patch does not hurt but i keep these #if 0 #endif just in case i have to revert things easily.
Revision
1.41 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Sun Jul 28 13:06:45 2002 UTC (21 years, 9 months ago) by
chl
Branch:
MAIN
Changes since
1.40: +2 -7 lines
Diff to
previous 1.40
Modified SKIP mb behaviour for Pframes when Bframes are active: Check if
intermediate B-block can be skipped, too. Check for SKIP is now done
in encode_pframe(), not in MBCoding().
Revision
1.40 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Sun Jul 28 02:55:41 2002 UTC (21 years, 9 months ago) by
chl
Branch:
MAIN
Changes since
1.39: +262 -126 lines
Diff to
previous 1.39
Interpolated and DirectMode motion estimation (DiamondSearch) for B-Frames.
Slow, but working, this might have been the final step to a complete
implemenation of b-frame encoding. Btw... it's 5.a.m so don't trust
this CVS entry!
Revision
1.39 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Sat Jul 27 23:47:01 2002 UTC (21 years, 9 months ago) by
chl
Branch:
MAIN
Changes since
1.38: +231 -45 lines
Diff to
previous 1.38
Finally Cut&Paste and Search&Replaced Square8_MainSearch, so PMV_USESQUARES8
is now functional. Also, some minor fixes.
Revision
1.30 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Wed Jul 10 20:11:05 2002 UTC (21 years, 9 months ago) by
chl
Branch:
MAIN
Changes since
1.29: +292 -8 lines
Diff to
previous 1.29
Preparations for block-based-interpolation ME: PMVfastIntSearch16 (pre-version)
Revision
1.26 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Sat Jul 6 17:04:57 2002 UTC (21 years, 9 months ago) by
chl
Branch:
MAIN
Changes since
1.25: +50 -173 lines
Diff to
previous 1.25
Minor modifications for supporting multithreaded ME.
Without _SMP symbol there should be no changes in behaviour.
Revision
1.18 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Wed May 22 12:42:25 2002 UTC (21 years, 11 months ago) by
chl
Branch:
MAIN
Changes since
1.17: +309 -22 lines
Diff to
previous 1.17
Added AdvDiamond_MainSearch by sysKIn which seems to be faster and better
than normal Diamond_MainSearch, at least for plain fullpel search. It can
be activated by PMV_ADVANCEDDIAMOND16/8 flag.
Revision
1.16 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Sat May 11 23:54:30 2002 UTC (21 years, 11 months ago) by
chl
Branch:
MAIN
Changes since
1.15: +31 -5 lines
Diff to
previous 1.15
Use INTER4V only if neighbours move different than current block.
Save quality, some speedup.
Revision
1.13 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Thu May 9 00:42:35 2002 UTC (21 years, 11 months ago) by
chl
Branch:
MAIN
Changes since
1.12: +124 -73 lines
Diff to
previous 1.12
Tests to not double check vectors in PMVfast16/8 (only in prediction, not during diamonds phase).
Minor speedup.
Revision
1.10 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Thu Apr 25 23:24:59 2002 UTC (22 years ago) by
chl
Branch:
MAIN
Changes since
1.9: +102 -46 lines
Diff to
previous 1.9
Fixed some minor issues in ME, changed calc_delta16/8 to non-linear function of quant
(from a paper on H.26L). Now inter4v finally gives some improvement.
Revision
1.2 -
(
view)
(
download)
(
annotate)
-
[select for diffs]
Thu Mar 28 20:57:25 2002 UTC (22 years, 1 month ago) by
edgomez
Branch:
MAIN
Changes since
1.1: +392 -384 lines
Diff to
previous 1.1
Big patch, use it with care (tests done with GNU/Linux) :
- Trivial warnings in encoder.c line 282 ret1 = ret 2 = 0;
- Lot of "New line at end of file missing" warnings fixed (gcc3)
- All 2d arrays have been turned into 1d arrays
To force alignment of matrices on stack, use DECLARE_ALIGNED_MATRIX
(read portab.h)
Win32 users should give feedback.
This form allows you to request diffs between any two revisions of this file.
For each of the two "sides" of the diff,
select a symbolic revision name using the selection box, or choose
'Use Text Field' and enter a numeric revision.
