1
0
mirror of https://github.com/FFmpeg/FFmpeg.git synced 2024-12-02 03:06:28 +02:00
FFmpeg/libavfilter/vf_remap.c

408 lines
18 KiB
C
Raw Normal View History

/*
* Copyright (c) 2016 Floris Sluiter
*
* 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
*/
/**
* @file
* Pixel remap filter
* This filter copies pixel by pixel a source frame to a target frame.
* It remaps the pixels to a new x,y destination based on two files ymap/xmap.
* Map files are passed as a parameter and are in PGM format (P2 or P5),
* where the values are y(rows)/x(cols) coordinates of the source_frame.
* The *target* frame dimension is based on mapfile dimensions: specified in the
* header of the mapfile and reflected in the number of datavalues.
* Dimensions of ymap and xmap must be equal. Datavalues must be positive or zero.
* Any datavalue in the ymap or xmap which value is higher
* then the *source* frame height or width is silently ignored, leaving a
* blank/chromakey pixel. This can safely be used as a feature to create overlays.
*
* Algorithm digest:
* Target_frame[y][x] = Source_frame[ ymap[y][x] ][ [xmap[y][x] ];
*/
#include "libavutil/colorspace.h"
#include "libavutil/imgutils.h"
#include "libavutil/pixdesc.h"
#include "libavutil/opt.h"
#include "avfilter.h"
#include "drawutils.h"
#include "formats.h"
2017-08-31 19:47:37 +02:00
#include "framesync.h"
#include "internal.h"
#include "video.h"
typedef struct RemapContext {
const AVClass *class;
int format;
int nb_planes;
int nb_components;
int step;
uint8_t fill_rgba[4];
int fill_color[4];
FFFrameSync fs;
int (*remap_slice)(AVFilterContext *ctx, void *arg, int jobnr, int nb_jobs);
} RemapContext;
#define OFFSET(x) offsetof(RemapContext, x)
#define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM
static const AVOption remap_options[] = {
{ "format", "set output format", OFFSET(format), AV_OPT_TYPE_INT, {.i64=0}, 0, 1, FLAGS, "format" },
{ "color", "", 0, AV_OPT_TYPE_CONST, {.i64=0}, .flags = FLAGS, .unit = "format" },
{ "gray", "", 0, AV_OPT_TYPE_CONST, {.i64=1}, .flags = FLAGS, .unit = "format" },
{ "fill", "set the color of the unmapped pixels", OFFSET(fill_rgba), AV_OPT_TYPE_COLOR, {.str="black"}, .flags = FLAGS },
{ NULL }
};
AVFILTER_DEFINE_CLASS(remap);
typedef struct ThreadData {
AVFrame *in, *xin, *yin, *out;
int nb_planes;
int nb_components;
int step;
} ThreadData;
static int query_formats(AVFilterContext *ctx)
{
RemapContext *s = ctx->priv;
static const enum AVPixelFormat pix_fmts[] = {
AV_PIX_FMT_YUVA444P,
AV_PIX_FMT_YUV444P,
AV_PIX_FMT_YUVJ444P,
AV_PIX_FMT_RGB24, AV_PIX_FMT_BGR24,
AV_PIX_FMT_ARGB, AV_PIX_FMT_ABGR, AV_PIX_FMT_RGBA, AV_PIX_FMT_BGRA,
AV_PIX_FMT_GBRP, AV_PIX_FMT_GBRAP,
2016-08-10 10:02:05 +02:00
AV_PIX_FMT_YUV444P9, AV_PIX_FMT_YUV444P10, AV_PIX_FMT_YUV444P12,
AV_PIX_FMT_YUV444P14, AV_PIX_FMT_YUV444P16,
AV_PIX_FMT_YUVA444P9, AV_PIX_FMT_YUVA444P10, AV_PIX_FMT_YUVA444P12, AV_PIX_FMT_YUVA444P16,
2016-08-10 10:02:05 +02:00
AV_PIX_FMT_GBRP9, AV_PIX_FMT_GBRP10, AV_PIX_FMT_GBRP12,
AV_PIX_FMT_GBRP14, AV_PIX_FMT_GBRP16,
AV_PIX_FMT_GBRAP10, AV_PIX_FMT_GBRAP12, AV_PIX_FMT_GBRAP16,
AV_PIX_FMT_RGB48, AV_PIX_FMT_BGR48,
AV_PIX_FMT_RGBA64, AV_PIX_FMT_BGRA64,
AV_PIX_FMT_NONE
};
static const enum AVPixelFormat gray_pix_fmts[] = {
AV_PIX_FMT_GRAY8, AV_PIX_FMT_GRAY9,
AV_PIX_FMT_GRAY10, AV_PIX_FMT_GRAY12,
AV_PIX_FMT_GRAY14, AV_PIX_FMT_GRAY16,
AV_PIX_FMT_NONE
};
static const enum AVPixelFormat map_fmts[] = {
AV_PIX_FMT_GRAY16,
AV_PIX_FMT_NONE
};
AVFilterFormats *pix_formats = NULL, *map_formats = NULL;
int ret;
pix_formats = ff_make_format_list(s->format ? gray_pix_fmts : pix_fmts);
if ((ret = ff_formats_ref(pix_formats, &ctx->inputs[0]->outcfg.formats)) < 0 ||
(ret = ff_formats_ref(pix_formats, &ctx->outputs[0]->incfg.formats)) < 0)
return ret;
map_formats = ff_make_format_list(map_fmts);
if ((ret = ff_formats_ref(map_formats, &ctx->inputs[1]->outcfg.formats)) < 0)
return ret;
return ff_formats_ref(map_formats, &ctx->inputs[2]->outcfg.formats);
}
/**
* remap_planar algorithm expects planes of same size
* pixels are copied from source to target using :
* Target_frame[y][x] = Source_frame[ ymap[y][x] ][ [xmap[y][x] ];
*/
2018-09-11 13:38:36 +02:00
#define DEFINE_REMAP_PLANAR_FUNC(name, bits, div) \
static int remap_planar##bits##_##name##_slice(AVFilterContext *ctx, void *arg, \
int jobnr, int nb_jobs) \
{ \
RemapContext *s = ctx->priv; \
const ThreadData *td = arg; \
2018-09-11 13:38:36 +02:00
const AVFrame *in = td->in; \
const AVFrame *xin = td->xin; \
const AVFrame *yin = td->yin; \
const AVFrame *out = td->out; \
const int slice_start = (out->height * jobnr ) / nb_jobs; \
const int slice_end = (out->height * (jobnr+1)) / nb_jobs; \
const int xlinesize = xin->linesize[0] / 2; \
const int ylinesize = yin->linesize[0] / 2; \
int x , y, plane; \
\
for (plane = 0; plane < td->nb_planes ; plane++) { \
const int dlinesize = out->linesize[plane] / div; \
const uint##bits##_t *src = (const uint##bits##_t *)in->data[plane]; \
uint##bits##_t *dst = (uint##bits##_t *)out->data[plane] + slice_start * dlinesize; \
const int slinesize = in->linesize[plane] / div; \
const uint16_t *xmap = (const uint16_t *)xin->data[0] + slice_start * xlinesize; \
const uint16_t *ymap = (const uint16_t *)yin->data[0] + slice_start * ylinesize; \
const int color = s->fill_color[plane]; \
2018-09-11 13:38:36 +02:00
\
for (y = slice_start; y < slice_end; y++) { \
for (x = 0; x < out->width; x++) { \
if (ymap[x] < in->height && xmap[x] < in->width) { \
dst[x] = src[ymap[x] * slinesize + xmap[x]]; \
} else { \
dst[x] = color; \
2018-09-11 13:38:36 +02:00
} \
} \
dst += dlinesize; \
xmap += xlinesize; \
ymap += ylinesize; \
} \
} \
\
return 0; \
}
2018-09-11 13:38:36 +02:00
DEFINE_REMAP_PLANAR_FUNC(nearest, 8, 1)
DEFINE_REMAP_PLANAR_FUNC(nearest, 16, 2)
2016-08-10 10:02:05 +02:00
/**
* remap_packed algorithm expects pixels with both padded bits (step) and
* number of components correctly set.
* pixels are copied from source to target using :
* Target_frame[y][x] = Source_frame[ ymap[y][x] ][ [xmap[y][x] ];
*/
2018-09-11 13:38:36 +02:00
#define DEFINE_REMAP_PACKED_FUNC(name, bits, div) \
static int remap_packed##bits##_##name##_slice(AVFilterContext *ctx, void *arg, \
int jobnr, int nb_jobs) \
{ \
RemapContext *s = ctx->priv; \
const ThreadData *td = arg; \
2018-09-11 13:38:36 +02:00
const AVFrame *in = td->in; \
const AVFrame *xin = td->xin; \
const AVFrame *yin = td->yin; \
const AVFrame *out = td->out; \
const int slice_start = (out->height * jobnr ) / nb_jobs; \
const int slice_end = (out->height * (jobnr+1)) / nb_jobs; \
const int dlinesize = out->linesize[0] / div; \
const int slinesize = in->linesize[0] / div; \
const int xlinesize = xin->linesize[0] / 2; \
const int ylinesize = yin->linesize[0] / 2; \
const uint##bits##_t *src = (const uint##bits##_t *)in->data[0]; \
uint##bits##_t *dst = (uint##bits##_t *)out->data[0] + slice_start * dlinesize; \
const uint16_t *xmap = (const uint16_t *)xin->data[0] + slice_start * xlinesize; \
const uint16_t *ymap = (const uint16_t *)yin->data[0] + slice_start * ylinesize; \
const int step = td->step / div; \
int c, x, y; \
\
for (y = slice_start; y < slice_end; y++) { \
for (x = 0; x < out->width; x++) { \
for (c = 0; c < td->nb_components; c++) { \
if (ymap[x] < in->height && xmap[x] < in->width) { \
dst[x * step + c] = src[ymap[x] * slinesize + xmap[x] * step + c]; \
} else { \
dst[x * step + c] = s->fill_color[c]; \
2018-09-11 13:38:36 +02:00
} \
} \
} \
dst += dlinesize; \
xmap += xlinesize; \
ymap += ylinesize; \
} \
\
return 0; \
}
2018-09-11 13:38:36 +02:00
DEFINE_REMAP_PACKED_FUNC(nearest, 8, 1)
DEFINE_REMAP_PACKED_FUNC(nearest, 16, 2)
2016-08-10 10:02:05 +02:00
static int config_input(AVFilterLink *inlink)
{
AVFilterContext *ctx = inlink->dst;
RemapContext *s = ctx->priv;
const AVPixFmtDescriptor *desc = av_pix_fmt_desc_get(inlink->format);
int depth = desc->comp[0].depth;
int is_rgb = !!(desc->flags & AV_PIX_FMT_FLAG_RGB);
int factor = 1 << (depth - 8);
uint8_t rgba_map[4];
ff_fill_rgba_map(rgba_map, inlink->format);
s->nb_planes = av_pix_fmt_count_planes(inlink->format);
s->nb_components = desc->nb_components;
if (is_rgb) {
s->fill_color[rgba_map[0]] = s->fill_rgba[0] * factor;
s->fill_color[rgba_map[1]] = s->fill_rgba[1] * factor;
s->fill_color[rgba_map[2]] = s->fill_rgba[2] * factor;
s->fill_color[rgba_map[3]] = s->fill_rgba[3] * factor;
} else {
s->fill_color[0] = RGB_TO_Y_BT709(s->fill_rgba[0], s->fill_rgba[1], s->fill_rgba[2]) * factor;
s->fill_color[1] = RGB_TO_U_BT709(s->fill_rgba[0], s->fill_rgba[1], s->fill_rgba[2], 0) * factor;
s->fill_color[2] = RGB_TO_V_BT709(s->fill_rgba[0], s->fill_rgba[1], s->fill_rgba[2], 0) * factor;
s->fill_color[3] = s->fill_rgba[3] * factor;
}
if (depth == 8) {
2016-08-10 10:02:05 +02:00
if (s->nb_planes > 1 || s->nb_components == 1) {
2018-09-11 13:38:36 +02:00
s->remap_slice = remap_planar8_nearest_slice;
2016-08-10 10:02:05 +02:00
} else {
2018-09-11 13:38:36 +02:00
s->remap_slice = remap_packed8_nearest_slice;
2016-08-10 10:02:05 +02:00
}
} else {
2016-08-10 10:02:05 +02:00
if (s->nb_planes > 1 || s->nb_components == 1) {
2018-09-11 13:38:36 +02:00
s->remap_slice = remap_planar16_nearest_slice;
2016-08-10 10:02:05 +02:00
} else {
2018-09-11 13:38:36 +02:00
s->remap_slice = remap_packed16_nearest_slice;
2016-08-10 10:02:05 +02:00
}
}
s->step = av_get_padded_bits_per_pixel(desc) >> 3;
return 0;
}
static int process_frame(FFFrameSync *fs)
{
AVFilterContext *ctx = fs->parent;
RemapContext *s = fs->opaque;
AVFilterLink *outlink = ctx->outputs[0];
AVFrame *out, *in, *xpic, *ypic;
int ret;
2017-08-31 19:47:37 +02:00
if ((ret = ff_framesync_get_frame(&s->fs, 0, &in, 0)) < 0 ||
(ret = ff_framesync_get_frame(&s->fs, 1, &xpic, 0)) < 0 ||
(ret = ff_framesync_get_frame(&s->fs, 2, &ypic, 0)) < 0)
return ret;
{
ThreadData td;
out = ff_get_video_buffer(outlink, outlink->w, outlink->h);
if (!out)
return AVERROR(ENOMEM);
av_frame_copy_props(out, in);
td.in = in;
td.xin = xpic;
td.yin = ypic;
td.out = out;
td.nb_planes = s->nb_planes;
td.nb_components = s->nb_components;
td.step = s->step;
ff_filter_execute(ctx, s->remap_slice, &td, NULL,
FFMIN(outlink->h, ff_filter_get_nb_threads(ctx)));
}
out->pts = av_rescale_q(s->fs.pts, s->fs.time_base, outlink->time_base);
return ff_filter_frame(outlink, out);
}
static int config_output(AVFilterLink *outlink)
{
AVFilterContext *ctx = outlink->src;
RemapContext *s = ctx->priv;
AVFilterLink *srclink = ctx->inputs[0];
AVFilterLink *xlink = ctx->inputs[1];
AVFilterLink *ylink = ctx->inputs[2];
FFFrameSyncIn *in;
int ret;
if (xlink->w != ylink->w || xlink->h != ylink->h) {
av_log(ctx, AV_LOG_ERROR, "Second input link %s parameters "
"(size %dx%d) do not match the corresponding "
"third input link %s parameters (%dx%d)\n",
ctx->input_pads[1].name, xlink->w, xlink->h,
ctx->input_pads[2].name, ylink->w, ylink->h);
return AVERROR(EINVAL);
}
outlink->w = xlink->w;
outlink->h = xlink->h;
outlink->sample_aspect_ratio = srclink->sample_aspect_ratio;
outlink->frame_rate = srclink->frame_rate;
2017-08-31 19:47:37 +02:00
ret = ff_framesync_init(&s->fs, ctx, 3);
if (ret < 0)
return ret;
in = s->fs.in;
in[0].time_base = srclink->time_base;
in[1].time_base = xlink->time_base;
in[2].time_base = ylink->time_base;
in[0].sync = 2;
in[0].before = EXT_STOP;
in[0].after = EXT_STOP;
in[1].sync = 1;
in[1].before = EXT_NULL;
in[1].after = EXT_INFINITY;
in[2].sync = 1;
in[2].before = EXT_NULL;
in[2].after = EXT_INFINITY;
s->fs.opaque = s;
s->fs.on_event = process_frame;
ret = ff_framesync_configure(&s->fs);
outlink->time_base = s->fs.time_base;
return ret;
}
static int activate(AVFilterContext *ctx)
{
RemapContext *s = ctx->priv;
2017-08-31 19:47:37 +02:00
return ff_framesync_activate(&s->fs);
}
static av_cold void uninit(AVFilterContext *ctx)
{
RemapContext *s = ctx->priv;
2017-08-31 19:47:37 +02:00
ff_framesync_uninit(&s->fs);
}
static const AVFilterPad remap_inputs[] = {
{
.name = "source",
.type = AVMEDIA_TYPE_VIDEO,
.config_props = config_input,
},
{
.name = "xmap",
.type = AVMEDIA_TYPE_VIDEO,
},
{
.name = "ymap",
.type = AVMEDIA_TYPE_VIDEO,
},
};
static const AVFilterPad remap_outputs[] = {
{
.name = "default",
.type = AVMEDIA_TYPE_VIDEO,
.config_props = config_output,
},
};
const AVFilter ff_vf_remap = {
.name = "remap",
.description = NULL_IF_CONFIG_SMALL("Remap pixels."),
.priv_size = sizeof(RemapContext),
.uninit = uninit,
.activate = activate,
2021-08-12 13:05:31 +02:00
FILTER_INPUTS(remap_inputs),
FILTER_OUTPUTS(remap_outputs),
avfilter: Replace query_formats callback with union of list and callback If one looks at the many query_formats callbacks in existence, one will immediately recognize that there is one type of default callback for video and a slightly different default callback for audio: It is "return ff_set_common_formats_from_list(ctx, pix_fmts);" for video with a filter-specific pix_fmts list. For audio, it is the same with a filter-specific sample_fmts list together with ff_set_common_all_samplerates() and ff_set_common_all_channel_counts(). This commit allows to remove the boilerplate query_formats callbacks by replacing said callback with a union consisting the old callback and pointers for pixel and sample format arrays. For the not uncommon case in which these lists only contain a single entry (besides the sentinel) enum AVPixelFormat and enum AVSampleFormat fields are also added to the union to store them directly in the AVFilter, thereby avoiding a relocation. The state of said union will be contained in a new, dedicated AVFilter field (the nb_inputs and nb_outputs fields have been shrunk to uint8_t in order to create a hole for this new field; this is no problem, as the maximum of all the nb_inputs is four; for nb_outputs it is only two). The state's default value coincides with the earlier default of query_formats being unset, namely that the filter accepts all formats (and also sample rates and channel counts/layouts for audio) provided that these properties agree coincide for all inputs and outputs. By using different union members for audio and video filters the type-unsafety of using the same functions for audio and video lists will furthermore be more confined to formats.c than before. When the new fields are used, they will also avoid allocations: Currently something nearly equivalent to ff_default_query_formats() is called after every successful call to a query_formats callback; yet in the common case that the newly allocated AVFilterFormats are not used at all (namely if there are no free links) these newly allocated AVFilterFormats are freed again without ever being used. Filters no longer using the callback will not exhibit this any more. Reviewed-by: Paul B Mahol <onemda@gmail.com> Reviewed-by: Nicolas George <george@nsup.org> Signed-off-by: Andreas Rheinhardt <andreas.rheinhardt@outlook.com>
2021-09-27 12:07:35 +02:00
FILTER_QUERY_FUNC(query_formats),
.priv_class = &remap_class,
.flags = AVFILTER_FLAG_SLICE_THREADS,
};