FFmpeg  4.0
avf_avectorscope.c
Go to the documentation of this file.
1 /*
2  * Copyright (c) 2013 Paul B Mahol
3  *
4  * This file is part of FFmpeg.
5  *
6  * FFmpeg is free software; you can redistribute it and/or
7  * modify it under the terms of the GNU Lesser General Public
8  * License as published by the Free Software Foundation; either
9  * version 2.1 of the License, or (at your option) any later version.
10  *
11  * FFmpeg is distributed in the hope that it will be useful,
12  * but WITHOUT ANY WARRANTY; without even the implied warranty of
13  * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
14  * Lesser General Public License for more details.
15  *
16  * You should have received a copy of the GNU Lesser General Public
17  * License along with FFmpeg; if not, write to the Free Software
18  * Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301 USA
19  */
20 
21 /**
22  * @file
23  * audio to video multimedia vectorscope filter
24  */
25 
26 #include "libavutil/avassert.h"
28 #include "libavutil/opt.h"
29 #include "libavutil/parseutils.h"
30 #include "avfilter.h"
31 #include "formats.h"
32 #include "audio.h"
33 #include "video.h"
34 #include "internal.h"
35 
41 };
42 
44  DOT,
47 };
48 
50  LIN,
53  LOG,
55 };
56 
57 typedef struct AudioVectorScopeContext {
58  const AVClass *class;
60  int w, h;
61  int hw, hh;
62  int mode;
63  int draw;
64  int scale;
65  int contrast[4];
66  int fade[4];
67  double zoom;
68  int swap;
69  int mirror;
70  unsigned prev_x, prev_y;
73 
74 #define OFFSET(x) offsetof(AudioVectorScopeContext, x)
75 #define FLAGS AV_OPT_FLAG_FILTERING_PARAM|AV_OPT_FLAG_VIDEO_PARAM
76 
77 static const AVOption avectorscope_options[] = {
78  { "mode", "set mode", OFFSET(mode), AV_OPT_TYPE_INT, {.i64=LISSAJOUS}, 0, MODE_NB-1, FLAGS, "mode" },
79  { "m", "set mode", OFFSET(mode), AV_OPT_TYPE_INT, {.i64=LISSAJOUS}, 0, MODE_NB-1, FLAGS, "mode" },
80  { "lissajous", "", 0, AV_OPT_TYPE_CONST, {.i64=LISSAJOUS}, 0, 0, FLAGS, "mode" },
81  { "lissajous_xy", "", 0, AV_OPT_TYPE_CONST, {.i64=LISSAJOUS_XY}, 0, 0, FLAGS, "mode" },
82  { "polar", "", 0, AV_OPT_TYPE_CONST, {.i64=POLAR}, 0, 0, FLAGS, "mode" },
83  { "rate", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str="25"}, 0, INT_MAX, FLAGS },
84  { "r", "set video rate", OFFSET(frame_rate), AV_OPT_TYPE_VIDEO_RATE, {.str="25"}, 0, INT_MAX, FLAGS },
85  { "size", "set video size", OFFSET(w), AV_OPT_TYPE_IMAGE_SIZE, {.str="400x400"}, 0, 0, FLAGS },
86  { "s", "set video size", OFFSET(w), AV_OPT_TYPE_IMAGE_SIZE, {.str="400x400"}, 0, 0, FLAGS },
87  { "rc", "set red contrast", OFFSET(contrast[0]), AV_OPT_TYPE_INT, {.i64=40}, 0, 255, FLAGS },
88  { "gc", "set green contrast", OFFSET(contrast[1]), AV_OPT_TYPE_INT, {.i64=160}, 0, 255, FLAGS },
89  { "bc", "set blue contrast", OFFSET(contrast[2]), AV_OPT_TYPE_INT, {.i64=80}, 0, 255, FLAGS },
90  { "ac", "set alpha contrast", OFFSET(contrast[3]), AV_OPT_TYPE_INT, {.i64=255}, 0, 255, FLAGS },
91  { "rf", "set red fade", OFFSET(fade[0]), AV_OPT_TYPE_INT, {.i64=15}, 0, 255, FLAGS },
92  { "gf", "set green fade", OFFSET(fade[1]), AV_OPT_TYPE_INT, {.i64=10}, 0, 255, FLAGS },
93  { "bf", "set blue fade", OFFSET(fade[2]), AV_OPT_TYPE_INT, {.i64=5}, 0, 255, FLAGS },
94  { "af", "set alpha fade", OFFSET(fade[3]), AV_OPT_TYPE_INT, {.i64=5}, 0, 255, FLAGS },
95  { "zoom", "set zoom factor", OFFSET(zoom), AV_OPT_TYPE_DOUBLE, {.dbl=1}, 0, 10, FLAGS },
96  { "draw", "set draw mode", OFFSET(draw), AV_OPT_TYPE_INT, {.i64=DOT}, 0, DRAW_NB-1, FLAGS, "draw" },
97  { "dot", "", 0, AV_OPT_TYPE_CONST, {.i64=DOT} , 0, 0, FLAGS, "draw" },
98  { "line", "", 0, AV_OPT_TYPE_CONST, {.i64=LINE}, 0, 0, FLAGS, "draw" },
99  { "scale", "set amplitude scale mode", OFFSET(scale), AV_OPT_TYPE_INT, {.i64=LIN}, 0, SCALE_NB-1, FLAGS, "scale" },
100  { "lin", "linear", 0, AV_OPT_TYPE_CONST, {.i64=LIN}, 0, 0, FLAGS, "scale" },
101  { "sqrt", "square root", 0, AV_OPT_TYPE_CONST, {.i64=SQRT}, 0, 0, FLAGS, "scale" },
102  { "cbrt", "cube root", 0, AV_OPT_TYPE_CONST, {.i64=CBRT}, 0, 0, FLAGS, "scale" },
103  { "log", "logarithmic", 0, AV_OPT_TYPE_CONST, {.i64=LOG}, 0, 0, FLAGS, "scale" },
104  { "swap", "swap x axis with y axis", OFFSET(swap), AV_OPT_TYPE_BOOL, {.i64=1}, 0, 1, FLAGS },
105  { "mirror", "mirror axis", OFFSET(mirror), AV_OPT_TYPE_INT, {.i64=2}, 0, 3, FLAGS, "mirror" },
106  { "none", "no mirror", 0, AV_OPT_TYPE_CONST, {.i64=0}, 0, 0, FLAGS, "mirror" },
107  { "x", "mirror x", 0, AV_OPT_TYPE_CONST, {.i64=1}, 0, 0, FLAGS, "mirror" },
108  { "y", "mirror y", 0, AV_OPT_TYPE_CONST, {.i64=2}, 0, 0, FLAGS, "mirror" },
109  { "xy", "mirror both", 0, AV_OPT_TYPE_CONST, {.i64=3}, 0, 0, FLAGS, "mirror" },
110  { NULL }
111 };
112 
113 AVFILTER_DEFINE_CLASS(avectorscope);
114 
115 static void draw_dot(AudioVectorScopeContext *s, unsigned x, unsigned y)
116 {
117  const int linesize = s->outpicref->linesize[0];
118  uint8_t *dst;
119 
120  if (s->zoom > 1) {
121  if (y >= s->h || x >= s->w)
122  return;
123  } else {
124  y = FFMIN(y, s->h - 1);
125  x = FFMIN(x, s->w - 1);
126  }
127 
128  dst = &s->outpicref->data[0][y * linesize + x * 4];
129  dst[0] = FFMIN(dst[0] + s->contrast[0], 255);
130  dst[1] = FFMIN(dst[1] + s->contrast[1], 255);
131  dst[2] = FFMIN(dst[2] + s->contrast[2], 255);
132  dst[3] = FFMIN(dst[3] + s->contrast[3], 255);
133 }
134 
135 static void draw_line(AudioVectorScopeContext *s, int x0, int y0, int x1, int y1)
136 {
137  int dx = FFABS(x1-x0), sx = x0 < x1 ? 1 : -1;
138  int dy = FFABS(y1-y0), sy = y0 < y1 ? 1 : -1;
139  int err = (dx>dy ? dx : -dy) / 2, e2;
140 
141  for (;;) {
142  draw_dot(s, x0, y0);
143 
144  if (x0 == x1 && y0 == y1)
145  break;
146 
147  e2 = err;
148 
149  if (e2 >-dx) {
150  err -= dy;
151  x0 += sx;
152  }
153 
154  if (e2 < dy) {
155  err += dx;
156  y0 += sy;
157  }
158  }
159 }
160 
162 {
163  const int linesize = s->outpicref->linesize[0];
164  int i, j;
165 
166  if (s->fade[0] || s->fade[1] || s->fade[2]) {
167  uint8_t *d = s->outpicref->data[0];
168  for (i = 0; i < s->h; i++) {
169  for (j = 0; j < s->w*4; j+=4) {
170  d[j+0] = FFMAX(d[j+0] - s->fade[0], 0);
171  d[j+1] = FFMAX(d[j+1] - s->fade[1], 0);
172  d[j+2] = FFMAX(d[j+2] - s->fade[2], 0);
173  d[j+3] = FFMAX(d[j+3] - s->fade[3], 0);
174  }
175  d += linesize;
176  }
177  }
178 }
179 
181 {
184  AVFilterLink *inlink = ctx->inputs[0];
185  AVFilterLink *outlink = ctx->outputs[0];
187  static const enum AVPixelFormat pix_fmts[] = { AV_PIX_FMT_RGBA, AV_PIX_FMT_NONE };
188  int ret;
189 
190  formats = ff_make_format_list(sample_fmts);
191  if ((ret = ff_formats_ref (formats, &inlink->out_formats )) < 0 ||
192  (ret = ff_add_channel_layout (&layout, AV_CH_LAYOUT_STEREO )) < 0 ||
193  (ret = ff_channel_layouts_ref (layout , &inlink->out_channel_layouts)) < 0)
194  return ret;
195 
196  formats = ff_all_samplerates();
197  if ((ret = ff_formats_ref(formats, &inlink->out_samplerates)) < 0)
198  return ret;
199 
200  formats = ff_make_format_list(pix_fmts);
201  if ((ret = ff_formats_ref(formats, &outlink->in_formats)) < 0)
202  return ret;
203 
204  return 0;
205 }
206 
207 static int config_input(AVFilterLink *inlink)
208 {
209  AVFilterContext *ctx = inlink->dst;
211  int nb_samples;
212 
213  nb_samples = FFMAX(1024, ((double)inlink->sample_rate / av_q2d(s->frame_rate)) + 0.5);
214  inlink->partial_buf_size =
215  inlink->min_samples =
216  inlink->max_samples = nb_samples;
217 
218  return 0;
219 }
220 
221 static int config_output(AVFilterLink *outlink)
222 {
223  AudioVectorScopeContext *s = outlink->src->priv;
224 
225  outlink->w = s->w;
226  outlink->h = s->h;
227  outlink->sample_aspect_ratio = (AVRational){1,1};
228  outlink->frame_rate = s->frame_rate;
229 
230  s->prev_x = s->hw = s->w / 2;
231  s->prev_y = s->hh = s->mode == POLAR ? s->h - 1 : s->h / 2;
232 
233  return 0;
234 }
235 
236 static int filter_frame(AVFilterLink *inlink, AVFrame *insamples)
237 {
238  AVFilterContext *ctx = inlink->dst;
239  AVFilterLink *outlink = ctx->outputs[0];
241  const int hw = s->hw;
242  const int hh = s->hh;
243  unsigned x, y;
244  unsigned prev_x = s->prev_x, prev_y = s->prev_y;
245  double zoom = s->zoom;
246  int i;
247 
248  if (!s->outpicref || s->outpicref->width != outlink->w ||
249  s->outpicref->height != outlink->h) {
251  s->outpicref = ff_get_video_buffer(outlink, outlink->w, outlink->h);
252  if (!s->outpicref) {
253  av_frame_free(&insamples);
254  return AVERROR(ENOMEM);
255  }
256 
258  for (i = 0; i < outlink->h; i++)
259  memset(s->outpicref->data[0] + i * s->outpicref->linesize[0], 0, outlink->w * 4);
260  }
261  s->outpicref->pts = insamples->pts;
262 
263  fade(s);
264 
265  if (zoom < 1) {
266  float max = 0;
267 
268  switch (insamples->format) {
269  case AV_SAMPLE_FMT_S16: {
270  int16_t *samples = (int16_t *)insamples->data[0];
271 
272  for (i = 0; i < insamples->nb_samples * 2; i++) {
273  float sample = samples[i] / (float)INT16_MAX;
274  max = FFMAX(FFABS(sample), max);
275  }
276 
277  }
278  break;
279  case AV_SAMPLE_FMT_FLT: {
280  float *samples = (float *)insamples->data[0];
281 
282  for (i = 0; i < insamples->nb_samples * 2; i++) {
283  max = FFMAX(FFABS(samples[i]), max);
284  }
285  }
286  break;
287  default:
288  av_assert2(0);
289  }
290 
291  zoom = 1. / max;
292  }
293 
294  for (i = 0; i < insamples->nb_samples; i++) {
295  int16_t *samples = (int16_t *)insamples->data[0] + i * 2;
296  float *samplesf = (float *)insamples->data[0] + i * 2;
297  float src[2];
298 
299  switch (insamples->format) {
300  case AV_SAMPLE_FMT_S16:
301  src[0] = samples[0] / (float)INT16_MAX;
302  src[1] = samples[1] / (float)INT16_MAX;
303  break;
304  case AV_SAMPLE_FMT_FLT:
305  src[0] = samplesf[0];
306  src[1] = samplesf[1];
307  break;
308  default:
309  av_assert2(0);
310  }
311 
312  switch (s->scale) {
313  case SQRT:
314  src[0] = FFSIGN(src[0]) * sqrtf(FFABS(src[0]));
315  src[1] = FFSIGN(src[1]) * sqrtf(FFABS(src[1]));
316  break;
317  case CBRT:
318  src[0] = FFSIGN(src[0]) * cbrtf(FFABS(src[0]));
319  src[1] = FFSIGN(src[1]) * cbrtf(FFABS(src[1]));
320  break;
321  case LOG:
322  src[0] = FFSIGN(src[0]) * logf(1 + FFABS(src[0])) / logf(2);
323  src[1] = FFSIGN(src[1]) * logf(1 + FFABS(src[1])) / logf(2);
324  break;
325  }
326 
327  if (s->mirror & 1)
328  src[0] = -src[0];
329 
330  if (s->mirror & 2)
331  src[1] = -src[1];
332 
333  if (s->swap)
334  FFSWAP(float, src[0], src[1]);
335 
336  if (s->mode == LISSAJOUS) {
337  x = ((src[1] - src[0]) * zoom / 2 + 1) * hw;
338  y = (1.0 - (src[0] + src[1]) * zoom / 2) * hh;
339  } else if (s->mode == LISSAJOUS_XY) {
340  x = (src[1] * zoom + 1) * hw;
341  y = (src[0] * zoom + 1) * hh;
342  } else {
343  float sx, sy, cx, cy;
344 
345  sx = src[1] * zoom;
346  sy = src[0] * zoom;
347  cx = sx * sqrtf(1 - 0.5 * sy * sy);
348  cy = sy * sqrtf(1 - 0.5 * sx * sx);
349  x = hw + hw * FFSIGN(cx + cy) * (cx - cy) * .7;
350  y = s->h - s->h * fabsf(cx + cy) * .7;
351  }
352 
353  if (s->draw == DOT) {
354  draw_dot(s, x, y);
355  } else {
356  draw_line(s, x, y, prev_x, prev_y);
357  }
358  prev_x = x;
359  prev_y = y;
360  }
361 
362  s->prev_x = x, s->prev_y = y;
363  av_frame_free(&insamples);
364 
365  return ff_filter_frame(outlink, av_frame_clone(s->outpicref));
366 }
367 
369 {
371 
373 }
374 
376  {
377  .name = "default",
378  .type = AVMEDIA_TYPE_AUDIO,
379  .config_props = config_input,
380  .filter_frame = filter_frame,
381  },
382  { NULL }
383 };
384 
386  {
387  .name = "default",
388  .type = AVMEDIA_TYPE_VIDEO,
389  .config_props = config_output,
390  },
391  { NULL }
392 };
393 
395  .name = "avectorscope",
396  .description = NULL_IF_CONFIG_SMALL("Convert input audio to vectorscope video output."),
397  .uninit = uninit,
398  .query_formats = query_formats,
399  .priv_size = sizeof(AudioVectorScopeContext),
400  .inputs = audiovectorscope_inputs,
401  .outputs = audiovectorscope_outputs,
402  .priv_class = &avectorscope_class,
403 };
#define NULL
Definition: coverity.c:32
const char * s
Definition: avisynth_c.h:768
This structure describes decoded (raw) audio or video data.
Definition: frame.h:218
AVOption.
Definition: opt.h:246
Main libavfilter public API header.
static void draw_line(AudioVectorScopeContext *s, int x0, int y0, int x1, int y1)
static av_cold void uninit(AVFilterContext *ctx)
AVFrame * ff_get_video_buffer(AVFilterLink *link, int w, int h)
Request a picture buffer with a specific set of permissions.
Definition: video.c:99
#define AV_CH_LAYOUT_STEREO
#define src
Definition: vp8dsp.c:254
#define sample
AVFilter ff_avf_avectorscope
AVFilterFormats * ff_make_format_list(const int *fmts)
Create a list of supported formats.
Definition: formats.c:283
const char * name
Pad name.
Definition: internal.h:60
AVFilterLink ** inputs
array of pointers to input links
Definition: avfilter.h:346
int ff_channel_layouts_ref(AVFilterChannelLayouts *f, AVFilterChannelLayouts **ref)
Add *ref as a new reference to f.
Definition: formats.c:435
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
Definition: avfilter.c:1080
uint8_t
#define av_cold
Definition: attributes.h:82
#define av_assert2(cond)
assert() equivalent, that does lie in speed critical code.
Definition: avassert.h:64
AVOptions.
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
Definition: frame.h:311
static const AVFilterPad audiovectorscope_outputs[]
static double av_q2d(AVRational a)
Convert an AVRational to a double.
Definition: rational.h:104
A filter pad used for either input or output.
Definition: internal.h:54
#define FLAGS
int width
Definition: frame.h:276
int ff_add_channel_layout(AVFilterChannelLayouts **l, uint64_t channel_layout)
Definition: formats.c:343
#define AVERROR(e)
Definition: error.h:43
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
Definition: frame.c:202
static int config_input(AVFilterLink *inlink)
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification. ...
Definition: internal.h:186
void * priv
private data for use by the filter
Definition: avfilter.h:353
simple assert() macros that are a bit more flexible than ISO C assert().
static const AVOption avectorscope_options[]
VectorScopeScale
#define FFMAX(a, b)
Definition: common.h:94
packed RGBA 8:8:8:8, 32bpp, RGBARGBA...
Definition: pixfmt.h:89
VectorScopeMode
audio channel layout utility functions
#define FFMIN(a, b)
Definition: common.h:96
int ff_formats_ref(AVFilterFormats *f, AVFilterFormats **ref)
Add *ref as a new reference to formats.
Definition: formats.c:440
#define FFSIGN(a)
Definition: common.h:73
AVFormatContext * ctx
Definition: movenc.c:48
#define FFABS(a)
Absolute value, Note, INT_MIN / INT64_MIN result in undefined behavior as they are not representable ...
Definition: common.h:72
static const AVFilterPad audiovectorscope_inputs[]
static void draw_dot(AudioVectorScopeContext *s, unsigned x, unsigned y)
static const AVFilterPad inputs[]
Definition: af_acontrast.c:193
AVFrame * av_frame_clone(const AVFrame *src)
Create a new frame that references the same data as src.
Definition: frame.c:538
static const AVFilterPad outputs[]
Definition: af_acontrast.c:203
AVFILTER_DEFINE_CLASS(avectorscope)
static int config_output(AVFilterLink *outlink)
A list of supported channel layouts.
Definition: formats.h:85
int format
format of the frame, -1 if unknown or unset Values correspond to enum AVPixelFormat for video frames...
Definition: frame.h:291
AVSampleFormat
Audio sample formats.
Definition: samplefmt.h:58
int linesize[AV_NUM_DATA_POINTERS]
For video, size in bytes of each picture line.
Definition: frame.h:249
AVRational sample_aspect_ratio
Sample aspect ratio for the video frame, 0/1 if unknown/unspecified.
Definition: frame.h:306
static av_always_inline float cbrtf(float x)
Definition: libm.h:61
Describe the class of an AVClass context structure.
Definition: log.h:67
Filter definition.
Definition: avfilter.h:144
Rational number (pair of numerator and denominator).
Definition: rational.h:58
offset must point to AVRational
Definition: opt.h:236
const char * name
Filter name.
Definition: avfilter.h:148
offset must point to two consecutive integers
Definition: opt.h:233
misc parsing utilities
AVFilterLink ** outputs
array of pointers to output links
Definition: avfilter.h:350
static enum AVPixelFormat pix_fmts[]
Definition: libkvazaar.c:266
AVFilterFormats * ff_all_samplerates(void)
Definition: formats.c:395
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Definition: frame.h:232
static int query_formats(AVFilterContext *ctx)
static int filter_frame(AVFilterLink *inlink, AVFrame *insamples)
signed 16 bits
Definition: samplefmt.h:61
A list of supported formats for one end of a filter link.
Definition: formats.h:64
uint64_t layout
VectorScopeDraw
An instance of a filter.
Definition: avfilter.h:338
static enum AVSampleFormat sample_fmts[]
Definition: adpcmenc.c:701
int height
Definition: frame.h:276
formats
Definition: signature.h:48
#define FFSWAP(type, a, b)
Definition: common.h:99
internal API functions
AVPixelFormat
Pixel format.
Definition: pixfmt.h:60
mode
Use these values in ebur128_init (or&#39;ed).
Definition: ebur128.h:83
int nb_samples
number of audio samples (per channel) described by this frame
Definition: frame.h:284
for(j=16;j >0;--j)
#define OFFSET(x)