falkTX
/
FFmpeg
mirror of https://github.com/falkTX/FFmpeg.git
1
0
Fork 0
Code Issues 0 Releases 338 Wiki Activity
Browse Source

lavfi: reimplement MPlayer's af_pan filter for libavfilter. 

Original code by Clément Bœsch.
Parameters parsing and misc enhancements by Nicolas George.
tags/n0.9
Clément Bœsch Nicolas George 14 years ago
parent
fd1cea6549
commit
1fbf7165d5
6 changed files with 359 additions and 2 deletions
Split View
Diff Options
Show Stats Download Patch File Download Diff File
  1. +1
    -0
      Changelog
  2. +48
    -0
      doc/filters.texi
  3. +1
    -0
      libavfilter/Makefile
  4. +306
    -0
      libavfilter/af_pan.c
  5. +1
    -0
      libavfilter/allfilters.c
  6. +2
    -2
      libavfilter/avfilter.h

+ 1
- 0
Changelog View File

@@ -122,6 +122,7 @@ easier to use. The changes are:
- VBLE Decoder
- OS X Video Decoder Acceleration (VDA) support
- compact and csv output in ffprobe
- pan audio filter


version 0.8:


+ 48
- 0
doc/filters.texi View File

@@ -235,6 +235,54 @@ the listener (standard for speakers).

Ported from SoX.

@section pan

Mix channels with specific gain levels. The filter accepts the output
channel layout followed by a set of channels definitions.

The filter accepts parameters of the form:
"@var{l}:@var{outdef}:@var{outdef}:..."

@table @option
@item l
output channel layout or number of channels

@item outdef
output channel specification, of the form:
"@var{out_name}=[@var{gain}*]@var{in_name}[+[@var{gain}*]@var{in_name}...]"

@item out_name
output channel to define, either a channel name (FL, FR, etc.) or a channel
number (c0, c1, etc.)

@item gain
multiplicative coefficient for the channel, 1 leaving the volume unchanged

@item in_name
input channel to use, see out_name for details; it is not possible to mix
named and numbered input channels
@end table

If the `=' in a channel specification is replaced by `<', then the gains for
that specification will be renormalized so that the total is 1, thus
avoiding clipping noise.

For example, if you want to down-mix from stereo to mono, but with a bigger
factor for the left channel:
@example
pan=1:c0=0.9*c0+0.1*c1
@end example

A customized down-mix to stereo that works automatically for 3-, 4-, 5- and
7-channels surround:
@example
pan=stereo: FL < FL + 0.5*FC + 0.6*BL + 0.6*SL : FR < FR + 0.5*FC + 0.6*BR + 0.6*SR
@end example

Note that @file{ffmpeg} integrates a default down-mix (and up-mix) system
that should be preferred (see "-ac" option) unless you have very specific
needs.

@section volume

Adjust the input audio volume.


+ 1
- 0
libavfilter/Makefile View File

@@ -29,6 +29,7 @@ OBJS-$(CONFIG_ANULL_FILTER) += af_anull.o
OBJS-$(CONFIG_ARESAMPLE_FILTER) += af_aresample.o
OBJS-$(CONFIG_ASHOWINFO_FILTER) += af_ashowinfo.o
OBJS-$(CONFIG_EARWAX_FILTER) += af_earwax.o
OBJS-$(CONFIG_PAN_FILTER) += af_pan.o
OBJS-$(CONFIG_VOLUME_FILTER) += af_volume.o

OBJS-$(CONFIG_ABUFFER_FILTER) += asrc_abuffer.o


+ 306
- 0
libavfilter/af_pan.c View File

@@ -0,0 +1,306 @@
/*
* Copyright (c) 2002 Anders Johansson <ajh@atri.curtin.edu.au>
* Copyright (c) 2011 Clément Bœsch <ubitux@gmail.com>
* Copyright (c) 2011 Nicolas George <nicolas.george@normalesup.org>
*
* 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 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
*/

/**
* @file
* Audio panning filter (channels mixing)
* Original code written by Anders Johansson for MPlayer,
* reimplemented for FFmpeg.
*/

#include <stdio.h>
#include "libavutil/audioconvert.h"
#include "libavutil/avstring.h"
#include "avfilter.h"
#include "internal.h"

#define MAX_CHANNELS 63

typedef struct {
int64_t out_channel_layout;
union {
double d[MAX_CHANNELS][MAX_CHANNELS];
// i is 1:7:8 fixed-point, i.e. in [-128*256; +128*256[
int i[MAX_CHANNELS][MAX_CHANNELS];
} gain;
int64_t need_renorm;
int need_renumber;
int nb_input_channels;
int nb_output_channels;
} PanContext;

static int parse_channel_name(char **arg, int *rchannel, int *rnamed)
{
char buf[8];
int len, i, channel_id;
int64_t layout, layout0;

if (sscanf(*arg, " %7[A-Z] %n", buf, &len)) {
layout0 = layout = av_get_channel_layout(buf);
for (i = 32; i > 0; i >>= 1) {
if (layout >= (int64_t)1 << i) {
channel_id += i;
layout >>= i;
}
}
if (channel_id >= MAX_CHANNELS || layout0 != (int64_t)1 << channel_id)
return AVERROR(EINVAL);
*rchannel = channel_id;
*rnamed = 1;
*arg += len;
return 0;
}
if (sscanf(*arg, " c%d %n", &channel_id, &len) &&
channel_id >= 0 && channel_id < MAX_CHANNELS) {
*rchannel = channel_id;
*rnamed = 0;
*arg += len;
return 0;
}
return AVERROR(EINVAL);
}

static void skip_spaces(char **arg)
{
int len = 0;

sscanf(*arg, " %n", &len);
*arg += len;
}

static av_cold int init(AVFilterContext *ctx, const char *args0, void *opaque)
{
PanContext *const pan = ctx->priv;
char *arg, *arg0, *tokenizer, *args = av_strdup(args0);
int out_ch_id, in_ch_id, len, named;
int nb_in_channels[2] = { 0, 0 }; // number of unnamed and named input channels
double gain;

if (!args)
return AVERROR(ENOMEM);
arg = av_strtok(args, ":", &tokenizer);
pan->out_channel_layout = av_get_channel_layout(arg);
if (!pan->out_channel_layout) {
av_log(ctx, AV_LOG_ERROR, "Unknown channel layout \"%s\"\n", arg);
return AVERROR(EINVAL);
}
pan->nb_output_channels = av_get_channel_layout_nb_channels(pan->out_channel_layout);

/* parse channel specifications */
while ((arg = arg0 = av_strtok(NULL, ":", &tokenizer))) {
/* channel name */
if (parse_channel_name(&arg, &out_ch_id, &named)) {
av_log(ctx, AV_LOG_ERROR,
"Expected out channel name, got \"%.8s\"\n", arg);
return AVERROR(EINVAL);
}
if (named) {
if (!((pan->out_channel_layout >> out_ch_id) & 1)) {
av_log(ctx, AV_LOG_ERROR,
"Channel \"%.8s\" does not exist in the chosen layout\n", arg0);
return AVERROR(EINVAL);
}
/* get the channel number in the output channel layout:
* out_channel_layout & ((1 << out_ch_id) - 1) are all the
* channels that come before out_ch_id,
* so their count is the index of out_ch_id */
out_ch_id = av_get_channel_layout_nb_channels(pan->out_channel_layout & (((int64_t)1 << out_ch_id) - 1));
}
if (out_ch_id < 0 || out_ch_id >= pan->nb_output_channels) {
av_log(ctx, AV_LOG_ERROR,
"Invalid out channel name \"%.8s\"\n", arg0);
return AVERROR(EINVAL);
}
if (*arg == '=') {
arg++;
} else if (*arg == '<') {
pan->need_renorm |= (int64_t)1 << out_ch_id;
arg++;
} else {
av_log(ctx, AV_LOG_ERROR,
"Syntax error after channel name in \"%.8s\"\n", arg0);
return AVERROR(EINVAL);
}
/* gains */
while (1) {
gain = 1;
if (sscanf(arg, " %lf %n* %n", &gain, &len, &len))
arg += len;
if (parse_channel_name(&arg, &in_ch_id, &named)){
av_log(ctx, AV_LOG_ERROR,
"Expected in channel name, got \"%.8s\"\n", arg);
return AVERROR(EINVAL);
}
nb_in_channels[named]++;
if (nb_in_channels[!named]) {
av_log(ctx, AV_LOG_ERROR,
"Can not mix named and numbered channels\n");
return AVERROR(EINVAL);
}
pan->gain.d[out_ch_id][in_ch_id] = gain;
if (!*arg)
break;
if (*arg != '+') {
av_log(ctx, AV_LOG_ERROR, "Syntax error near \"%.8s\"\n", arg);
return AVERROR(EINVAL);
}
arg++;
skip_spaces(&arg);
}
}
pan->need_renumber = !!nb_in_channels[1];

av_free(args);
return 0;
}

static int query_formats(AVFilterContext *ctx)
{
PanContext *pan = ctx->priv;
AVFilterLink *inlink = ctx->inputs[0];
AVFilterLink *outlink = ctx->outputs[0];
AVFilterFormats *formats;

const enum AVSampleFormat sample_fmts[] = {AV_SAMPLE_FMT_S16, -1};
const int packing_fmts[] = {AVFILTER_PACKED, -1};

avfilter_set_common_sample_formats (ctx, avfilter_make_format_list(sample_fmts));
avfilter_set_common_packing_formats(ctx, avfilter_make_format_list(packing_fmts));

// inlink supports any channel layout
formats = avfilter_make_all_channel_layouts();
avfilter_formats_ref(formats, &inlink->out_chlayouts);

// outlink supports only requested output channel layout
formats = NULL;
avfilter_add_format(&formats, pan->out_channel_layout);
avfilter_formats_ref(formats, &outlink->in_chlayouts);
return 0;
}

static int config_props(AVFilterLink *link)
{
AVFilterContext *ctx = link->dst;
PanContext *pan = ctx->priv;
char buf[1024], *cur;
int i, j, k, r;
double t;

pan->nb_input_channels = av_get_channel_layout_nb_channels(link->channel_layout);
if (pan->need_renumber) {
// input channels were given by their name: renumber them
for (i = j = 0; i < MAX_CHANNELS; i++) {
if ((link->channel_layout >> i) & 1) {
for (k = 0; k < pan->nb_output_channels; k++)
pan->gain.d[k][j] = pan->gain.d[k][i];
j++;
}
}
}
// renormalize
for (i = 0; i < pan->nb_output_channels; i++) {
if (!((pan->need_renorm >> i) & 1))
continue;
t = 0;
for (j = 0; j < pan->nb_input_channels; j++)
t += pan->gain.d[i][j];
if (t > -1E-5 && t < 1E-5) {
// t is almost 0 but not exactly, this is probably a mistake
if (t)
av_log(ctx, AV_LOG_WARNING,
"Degenerate coefficients while renormalizing\n");
continue;
}
for (j = 0; j < pan->nb_input_channels; j++)
pan->gain.d[i][j] /= t;
}
// summary
for (i = 0; i < pan->nb_output_channels; i++) {
cur = buf;
for (j = 0; j < pan->nb_input_channels; j++) {
r = snprintf(cur, buf + sizeof(buf) - cur, "%s%.3g i%d",
j ? " + " : "", pan->gain.d[i][j], j);
cur += FFMIN(buf + sizeof(buf) - cur, r);
}
av_log(ctx, AV_LOG_INFO, "o%d = %s\n", i, buf);
}
// convert to integer
for (i = 0; i < pan->nb_output_channels; i++) {
for (j = 0; j < pan->nb_input_channels; j++) {
if (pan->gain.d[i][j] < -128 || pan->gain.d[i][j] > 128)
av_log(ctx, AV_LOG_WARNING,
"Gain #%d->#%d too large, clamped\n", j, i);
pan->gain.i[i][j] = av_clipf(pan->gain.d[i][j], -128, 128) * 256.0;
}
}
return 0;
}


static void filter_samples(AVFilterLink *inlink, AVFilterBufferRef *insamples)
{
PanContext *const pan = inlink->dst->priv;
int i, o, n = insamples->audio->nb_samples;

/* input */
const int16_t *in = (int16_t *)insamples->data[0];
const int16_t *in_end = in + n * pan->nb_input_channels;

/* output */
AVFilterLink *const outlink = inlink->dst->outputs[0];
AVFilterBufferRef *outsamples = avfilter_get_audio_buffer(outlink, AV_PERM_WRITE, n);
int16_t *out = (int16_t *)outsamples->data[0];

for (; in < in_end; in += pan->nb_input_channels) {
for (o = 0; o < pan->nb_output_channels; o++) {
int v = 0;
for (i = 0; i < pan->nb_input_channels; i++)
v += pan->gain.i[o][i] * in[i];
*(out++) = v >> 8;
}
}

avfilter_filter_samples(outlink, outsamples);
avfilter_unref_buffer(insamples);
}

AVFilter avfilter_af_pan = {
.name = "pan",
.description = NULL_IF_CONFIG_SMALL("Remix channels with coefficients (panning)"),
.priv_size = sizeof(PanContext),
.init = init,
.query_formats = query_formats,

.inputs = (const AVFilterPad[]) {
{ .name = "default",
.type = AVMEDIA_TYPE_AUDIO,
.config_props = config_props,
.filter_samples = filter_samples,
.min_perms = AV_PERM_READ, },
{ .name = NULL}
},
.outputs = (const AVFilterPad[]) {
{ .name = "default",
.type = AVMEDIA_TYPE_AUDIO, },
{ .name = NULL}
},
};

+ 1
- 0
libavfilter/allfilters.c View File

@@ -40,6 +40,7 @@ void avfilter_register_all(void)
REGISTER_FILTER (ARESAMPLE, aresample, af);
REGISTER_FILTER (ASHOWINFO, ashowinfo, af);
REGISTER_FILTER (EARWAX, earwax, af);
REGISTER_FILTER (PAN, pan, af);
REGISTER_FILTER (VOLUME, volume, af);

REGISTER_FILTER (ABUFFER, abuffer, asrc);


+ 2
- 2
libavfilter/avfilter.h View File

@@ -29,8 +29,8 @@
#include "libavutil/rational.h"

#define LIBAVFILTER_VERSION_MAJOR 2
#define LIBAVFILTER_VERSION_MINOR 48
#define LIBAVFILTER_VERSION_MICRO 1
#define LIBAVFILTER_VERSION_MINOR 49
#define LIBAVFILTER_VERSION_MICRO 0

#define LIBAVFILTER_VERSION_INT AV_VERSION_INT(LIBAVFILTER_VERSION_MAJOR, \
LIBAVFILTER_VERSION_MINOR, \


Write Preview
Loading…
Cancel
Save