[FFmpeg-devel] [PATCH] avfilter: port qp filter from libmpcodecs
Michael Niedermayer
michaelni at gmx.at
Thu Sep 19 19:47:26 CEST 2013
On Thu, Sep 19, 2013 at 10:49:41AM +0000, Paul B Mahol wrote:
> On 9/19/13, Michael Niedermayer <michaelni at gmx.at> wrote:
> > On Wed, Sep 18, 2013 at 11:57:55PM +0000, Paul B Mahol wrote:
> >> Signed-off-by: Paul B Mahol <onemda at gmail.com>
> >> ---
> >> doc/filters.texi | 32 ++++++++++
> >> libavfilter/Makefile | 1 +
> >> libavfilter/allfilters.c | 1 +
> >> libavfilter/vf_qp.c | 152
> >> +++++++++++++++++++++++++++++++++++++++++++++++
> >> 4 files changed, 186 insertions(+)
> >> create mode 100644 libavfilter/vf_qp.c
> >>
> >> diff --git a/doc/filters.texi b/doc/filters.texi
> >> index 915f310..127885a 100644
> >> --- a/doc/filters.texi
> >> +++ b/doc/filters.texi
> >> @@ -6236,6 +6236,38 @@ The main purpose of setting @option{mp} to a chroma
> >> plane is to reduce CPU
> >> load and make pullup usable in realtime on slow machines.
> >> @end table
> >>
> >> + at section qp
> >> +
> >> +Change video quantization parameters (QP).
> >> +
> >> +The filter accepts the following option:
> >> +
> >> + at table @option
> >> + at item qp
> >> +Set expression for quantization parameter.
> >> + at end table
> >> +
> >> +The expression is evaluated through the eval API and can contain, among
> >> others,
> >> +the following constants:
> >> +
> >> + at table @var
> >> + at item known
> >> +1 if index is not 129, 0 otherwise.
> >> +
> >> + at item qp
> >> +Sequentional index starting from -129 to 128.
> >> + at end table
> >> +
> >> + at subsection Examples
> >> +
> >> + at itemize
> >> + at item
> >> +Some equation like:
> >> + at example
> >> +qp=2+2*sin(PI*qp)
> >> + at end example
> >> + at end itemize
> >> +
> >> @section removelogo
> >>
> >> Suppress a TV station logo, using an image file to determine which
> >> diff --git a/libavfilter/Makefile b/libavfilter/Makefile
> >> index 198bf4c..5dc8389 100644
> >> --- a/libavfilter/Makefile
> >> +++ b/libavfilter/Makefile
> >> @@ -174,6 +174,7 @@ OBJS-$(CONFIG_PIXDESCTEST_FILTER) +=
> >> vf_pixdesctest.o
> >> OBJS-$(CONFIG_PP_FILTER) += vf_pp.o
> >> OBJS-$(CONFIG_PSNR_FILTER) += vf_psnr.o dualinput.o
> >> OBJS-$(CONFIG_PULLUP_FILTER) += vf_pullup.o
> >> +OBJS-$(CONFIG_QP_FILTER) += vf_qp.o
> >> OBJS-$(CONFIG_REMOVELOGO_FILTER) += bbox.o lswsutils.o
> >> lavfutils.o vf_removelogo.o
> >> OBJS-$(CONFIG_ROTATE_FILTER) += vf_rotate.o
> >> OBJS-$(CONFIG_SEPARATEFIELDS_FILTER) += vf_separatefields.o
> >> diff --git a/libavfilter/allfilters.c b/libavfilter/allfilters.c
> >> index 2825304..69e6cab 100644
> >> --- a/libavfilter/allfilters.c
> >> +++ b/libavfilter/allfilters.c
> >> @@ -169,6 +169,7 @@ void avfilter_register_all(void)
> >> REGISTER_FILTER(PP, pp, vf);
> >> REGISTER_FILTER(PSNR, psnr, vf);
> >> REGISTER_FILTER(PULLUP, pullup, vf);
> >> + REGISTER_FILTER(QP, qp, vf);
> >> REGISTER_FILTER(REMOVELOGO, removelogo, vf);
> >> REGISTER_FILTER(ROTATE, rotate, vf);
> >> REGISTER_FILTER(SAB, sab, vf);
> >> diff --git a/libavfilter/vf_qp.c b/libavfilter/vf_qp.c
> >> new file mode 100644
> >> index 0000000..67fe2ec
> >> --- /dev/null
> >> +++ b/libavfilter/vf_qp.c
> >> @@ -0,0 +1,152 @@
> >> +/*
> >> + * Copyright (C) 2004 Michael Niedermayer <michaelni at gmx.at>
> >> + *
> >> + * This file is part of FFmpeg.
> >> + *
> >> + * FFmpeg is free software; you can redistribute it and/or
> >> + * modify it under the terms of the GNU Lesser General Public
> >> + * License as published by the Free Software Foundation; either
> >> + * version 2.1 of the License, or (at your option) any later version.
> >> + *
> >> + * FFmpeg is distributed in the hope that it will be useful,
> >> + * but WITHOUT ANY WARRANTY; without even the implied warranty of
> >> + * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
> >> + * Lesser General Public License for more details.
> >> + *
> >> + * You should have received a copy of the GNU Lesser General Public
> >> + * License along with FFmpeg; if not, write to the Free Software
> >> + * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA
> >> 02110-1301 USA
> >> + */
> >> +
> >> +#include <math.h>
> >> +#include "libavutil/eval.h"
> >> +#include "libavutil/imgutils.h"
> >> +#include "libavutil/pixdesc.h"
> >> +#include "libavutil/opt.h"
> >> +#include "avfilter.h"
> >> +#include "formats.h"
> >> +#include "internal.h"
> >> +#include "video.h"
> >> +
> >> +typedef struct QPContext {
> >> + const AVClass *class;
> >> + char *qp_expr_str;
> >> + int8_t lut[257];
> >> + int h, qstride;
> >> +} QPContext;
> >> +
> >> +#define OFFSET(x) offsetof(QPContext, x)
> >> +#define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM
> >> +
> >> +static const AVOption qp_options[] = {
> >> + { "qp", "set qp expression", OFFSET(qp_expr_str), AV_OPT_TYPE_STRING,
> >> {.str=NULL}, 0, 0, FLAGS },
> >> + { NULL }
> >> +};
> >> +
> >> +AVFILTER_DEFINE_CLASS(qp);
> >> +
> >> +static int config_input(AVFilterLink *inlink)
> >> +{
> >> + AVFilterContext *ctx = inlink->dst;
> >> + QPContext *s = ctx->priv;
> >> + int i;
> >> +
> >> + if (!s->qp_expr_str)
> >> + return 0;
> >> +
> >> + s->h = (inlink->h + 15) >> 4;
> >> + s->qstride = (inlink->w + 15) >> 4;
> >> + for (i = -129; i < 128; i++) {
> >> + double var_values[] = { i != -129, i, 0 };
> >> + static const char *var_names[] = { "known", "qp", NULL };
> >> + double temp_val;
> >> + int ret;
> >> +
> >> + ret = av_expr_parse_and_eval(&temp_val, s->qp_expr_str,
> >> + var_names, var_values,
> >> + NULL, NULL, NULL, NULL, 0, 0, ctx);
> >> + if (ret < 0)
> >> + return ret;
> >> +
> >> + s->lut[i + 129] = lrintf(temp_val);
> >> + }
> >
> > it could make sense to allow forcing the evaluating the expression for
> > each macroblock
> > that way it could be used to do spatial or temporal smoothing,
> > depend on frame number of do add a random value
>
> Gread idea, but becaues I can't compare it with broken mp=qp, I want to know
> is current code (use of API) correct?
>
> This patch is not about adding new (non-trivial/easy) features, one
> can do it freely later.
the API use should be ok but i dont think you need to clone the frame
[...]
--
Michael GnuPG fingerprint: 9FF2128B147EF6730BADF133611EC787040B0FAB
Avoid a single point of failure, be that a person or equipment.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://ffmpeg.org/pipermail/ffmpeg-devel/attachments/20130919/40dd177b/attachment.asc>
More information about the ffmpeg-devel
mailing list