FFmpeg: libavfilter/vf_deflicker.c Source File
Go to the documentation of this file. 1 /*
2 * Copyright (c) 2017 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
26
27 #define FF_BUFQUEUE_SIZE 129
29
33
34 #define SIZE FF_BUFQUEUE_SIZE
35
45 };
46
49
53
59
63
66
70 uint8_t *dst, ptrdiff_t dst_linesize,
int w,
int h,
float f);
72
73 #define OFFSET(x) offsetof(DeflickerContext, x)
74 #define FLAGS AV_OPT_FLAG_VIDEO_PARAM|AV_OPT_FLAG_FILTERING_PARAM
75
90 };
91
93
115 };
116
118 const uint8_t *
src, ptrdiff_t src_linesize,
119 uint8_t *dst, ptrdiff_t dst_linesize,
120 int w,
int h,
float f)
121 {
122 int x, y;
123
124 for (y = 0; y <
h; y++) {
125 for (x = 0; x <
w; x++) {
127 }
128
129 dst += dst_linesize;
131 }
132
133 return 0;
134 }
135
137 const uint8_t *ssrc, ptrdiff_t src_linesize,
138 uint8_t *ddst, ptrdiff_t dst_linesize,
139 int w,
int h,
float f)
140 {
142 const uint16_t *
src = (
const uint16_t *)ssrc;
143 uint16_t *dst = (uint16_t *)ddst;
144 const int max = (1 <<
s->depth) - 1;
145 int x, y;
146
147 for (y = 0; y <
h; y++) {
148 for (x = 0; x <
w; x++) {
150 }
151
152 dst += dst_linesize / 2;
153 src += src_linesize / 2;
154 }
155
156 return 0;
157 }
158
160 {
162 const uint8_t *
src = in->
data[0];
163 int64_t sum = 0;
164 int y, x;
165
166 memset(
s->histogram, 0, (1 <<
s->depth) *
sizeof(*
s->histogram));
167
168 for (y = 0; y <
s->planeheight[0]; y++) {
169 for (x = 0; x <
s->planewidth[0]; x++) {
170 s->histogram[
src[x]]++;
171 }
173 }
174
175 for (y = 0; y < 1 <<
s->depth; y++) {
176 sum +=
s->histogram[y] * y;
177 }
178
179 return 1.0f * sum / (
s->planeheight[0] *
s->planewidth[0]);
180 }
181
183 {
185 const uint16_t *
src = (
const uint16_t *)in->
data[0];
186 int64_t sum = 0;
187 int y, x;
188
189 memset(
s->histogram, 0, (1 <<
s->depth) *
sizeof(*
s->histogram));
190
191 for (y = 0; y <
s->planeheight[0]; y++) {
192 for (x = 0; x <
s->planewidth[0]; x++) {
193 s->histogram[
src[x]]++;
194 }
196 }
197
198 for (y = 0; y < 1 <<
s->depth; y++) {
199 sum +=
s->histogram[y] * y;
200 }
201
202 return 1.0f * sum / (
s->planeheight[0] *
s->planewidth[0]);
203 }
204
206 {
208 int y;
209
211
212 for (y = 0; y <
s->size; y++) {
213 *
f +=
s->luminance[y];
214 }
215
217 *
f /=
s->luminance[0];
218 }
219
221 {
223 int y;
224
226
227 for (y = 0; y <
s->size; y++) {
228 *
f *=
s->luminance[y];
229 }
230
231 *
f = pow(*
f, 1.0
f /
s->size);
232 *
f /=
s->luminance[0];
233 }
234
236 {
238 int y;
239
241
242 for (y = 0; y <
s->size; y++) {
243 *
f += 1.0f /
s->luminance[y];
244 }
245
247 *
f /=
s->luminance[0];
248 }
249
251 {
253 int y;
254
256
257 for (y = 0; y <
s->size; y++) {
258 *
f +=
s->luminance[y] *
s->luminance[y];
259 }
260
263 *
f /=
s->luminance[0];
264 }
265
267 {
269 int y;
270
272
273 for (y = 0; y <
s->size; y++) {
274 *
f +=
s->luminance[y] *
s->luminance[y] *
s->luminance[y];
275 }
276
279 *
f /=
s->luminance[0];
280 }
281
283 {
285 int y;
286
288
289 for (y = 0; y <
s->size; y++) {
290 *
f +=
powf(
s->luminance[y],
s->size);
291 }
292
295 *
f /=
s->luminance[0];
296 }
297
299 {
300 const float *aa =
a, *bb =
b;
301 return round(aa - bb);
302 }
303
305 {
307
308 memcpy(
s->sorted,
s->luminance,
sizeof(
s->sorted));
310
311 *
f =
s->sorted[
s->size >> 1] /
s->luminance[0];
312 }
313
315 {
319
320 s->nb_planes =
desc->nb_components;
321
323 s->planeheight[0] =
s->planeheight[3] =
inlink->h;
325 s->planewidth[0] =
s->planewidth[3] =
inlink->w;
326
327 s->depth =
desc->comp[0].depth;
331 } else {
334 }
335
336 s->histogram =
av_calloc(1 <<
s->depth,
sizeof(*
s->histogram));
339
348 }
349
350 return 0;
351 }
352
354 {
361 int y;
362
363 if (
s->q.available <
s->size && !
s->eof) {
364 s->luminance[
s->available] =
s->calc_avgy(
ctx, buf);
367 return 0;
368 }
369
371
376 }
377
378 s->get_factor(
ctx, &
f);
381 outlink->
w, outlink->
h,
f);
382 for (y = 1 -
s->bypass; y < s->nb_planes; y++) {
385 s->planewidth[y] * (1 + (
s->depth > 8)),
s->planeheight[y]);
386 }
387
389 metadata = &
out->metadata;
390 if (metadata) {
392
395
398
401 }
402
405 memmove(&
s->luminance[0], &
s->luminance[1],
sizeof(*
s->luminance) * (
s->size - 1));
406 s->luminance[
s->available - 1] =
s->calc_avgy(
ctx, buf);
408
410 }
411
413 {
417
421 if (!buf)
424 if (!buf)
426
430 }
431
433 }
434
436 {
438
441 }
442
444 {
449 },
450 };
451
453 {
457 },
458 };
459
464 .priv_class = &deflicker_class,
469 };
AVFrame * ff_get_video_buffer(AVFilterLink *link, int w, int h)
Request a picture buffer with a specific set of permissions.
#define AV_PIX_FMT_YUVA422P16
AVPixelFormat
Pixel format.
static void get_gm_factor(AVFilterContext *ctx, float *f)
Filter the word "frame" indicates either a video frame or a group of audio as stored in an AVFrame structure Format for each input and each output the list of supported formats For video that means pixel format For audio that means channel sample they are references to shared objects When the negotiation mechanism computes the intersection of the formats supported at each end of a all references to both lists are replaced with a reference to the intersection And when a single format is eventually chosen for a link amongst the remaining all references to the list are updated That means that if a filter requires that its input and output have the same format amongst a supported all it has to do is use a reference to the same list of formats query_formats can leave some formats unset and return AVERROR(EAGAIN) to cause the negotiation mechanism toagain later. That can be used by filters with complex requirements to use the format negotiated on one link to set the formats supported on another. Frame references ownership and permissions
static void get_am_factor(AVFilterContext *ctx, float *f)
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
const AVPixFmtDescriptor * av_pix_fmt_desc_get(enum AVPixelFormat pix_fmt)
#define AVERROR_EOF
End of file.
#define FILTER_PIXFMTS_ARRAY(array)
The exact code depends on how similar the blocks are and how related they are to the and needs to apply these operations to the correct inlink or outlink if there are several Macros are available to factor that when no extra processing is inlink
static int comparef(const void *a, const void *b)
void av_frame_free(AVFrame **frame)
Free the frame and any dynamically allocated objects in it, e.g.
#define AV_PIX_FMT_YUVA422P9
This structure describes decoded (raw) audio or video data.
#define AV_PIX_FMT_YUVA420P16
#define AV_PIX_FMT_YUVA420P10
#define AV_PIX_FMT_YUV420P10
int ff_request_frame(AVFilterLink *link)
Request an input frame from the filter at the other end of the link.
static const AVOption deflicker_options[]
@ AV_PIX_FMT_YUV440P
planar YUV 4:4:0 (1 Cr & Cb sample per 1x2 Y samples)
const char * name
Filter name.
A link between two filters.
#define AV_PIX_FMT_YUVA422P10
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
static AVFrame * ff_bufqueue_get(struct FFBufQueue *queue)
Get the first buffer from the queue and remove it.
void av_image_copy_plane(uint8_t *dst, int dst_linesize, const uint8_t *src, int src_linesize, int bytewidth, int height)
Copy image plane from src to dst.
#define AV_PIX_FMT_YUVA420P9
static void get_pm_factor(AVFilterContext *ctx, float *f)
float(* calc_avgy)(AVFilterContext *ctx, AVFrame *in)
#define AV_PIX_FMT_YUVA444P16
#define AV_PIX_FMT_YUV422P9
void(* get_factor)(AVFilterContext *ctx, float *f)
#define AV_PIX_FMT_GRAY16
A filter pad used for either input or output.
#define AV_PIX_FMT_YUV444P10
@ AV_PIX_FMT_YUVJ411P
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples) full scale (JPEG), deprecated in favor ...
static int deflicker16(AVFilterContext *ctx, const uint8_t *ssrc, ptrdiff_t src_linesize, uint8_t *ddst, ptrdiff_t dst_linesize, int w, int h, float f)
#define AV_PIX_FMT_YUV422P16
@ AV_PIX_FMT_YUVJ422P
planar YUV 4:2:2, 16bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV422P and setting col...
@ AV_PIX_FMT_YUVA420P
planar YUV 4:2:0, 20bpp, (1 Cr & Cb sample per 2x2 Y & A samples)
#define AV_PIX_FMT_YUV444P16
#define AV_CEIL_RSHIFT(a, b)
static float calc_avgy8(AVFilterContext *ctx, AVFrame *in)
#define AV_PIX_FMT_YUVA444P12
#define AV_PIX_FMT_YUV420P9
#define AV_PIX_FMT_YUV420P16
#define AV_PIX_FMT_GRAY14
AVFrame * av_frame_clone(const AVFrame *src)
Create a new frame that references the same data as src.
@ AV_PIX_FMT_YUV420P
planar YUV 4:2:0, 12bpp, (1 Cr & Cb sample per 2x2 Y samples)
#define FILTER_INPUTS(array)
@ AV_PIX_FMT_YUVJ444P
planar YUV 4:4:4, 24bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV444P and setting col...
#define AV_PIX_FMT_GRAY10
Describe the class of an AVClass context structure.
int av_frame_copy_props(AVFrame *dst, const AVFrame *src)
Copy only "metadata" fields from src to dst.
const AVFilter ff_vf_deflicker
@ AV_PIX_FMT_YUVJ420P
planar YUV 4:2:0, 12bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV420P and setting col...
static void ff_bufqueue_discard_all(struct FFBufQueue *queue)
Unref and remove all buffers from the queue.
#define AV_PIX_FMT_YUV440P10
static __device__ float sqrtf(float a)
#define AV_PIX_FMT_YUV422P10
@ AV_PIX_FMT_GRAY8
Y , 8bpp.
static const AVFilterPad outputs[]
static int request_frame(AVFilterLink *outlink)
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
#define AV_PIX_FMT_YUV422P12
static const AVFilterPad inputs[]
#define AV_PIX_FMT_YUV444P12
static av_cold void uninit(AVFilterContext *ctx)
AVFilterContext * src
source filter
The reader does not expect b to be semantically here and if the code is changed by maybe adding a a division or other the signedness will almost certainly be mistaken To avoid this confusion a new type was SUINT is the C unsigned type but it holds a signed int to use the same example SUINT a
@ AV_PIX_FMT_YUVA444P
planar YUV 4:4:4 32bpp, (1 Cr & Cb sample per 1x1 Y & A samples)
static int deflicker8(AVFilterContext *ctx, const uint8_t *src, ptrdiff_t src_linesize, uint8_t *dst, ptrdiff_t dst_linesize, int w, int h, float f)
#define AV_PIX_FMT_YUVA444P10
static void ff_bufqueue_add(void *log, struct FFBufQueue *queue, AVFrame *buf)
Add a buffer to the queue.
static int filter_frame(AVFilterLink *inlink, AVFrame *buf)
AVFILTER_DEFINE_CLASS(deflicker)
static void get_median_factor(AVFilterContext *ctx, float *f)
static AVFrame * ff_bufqueue_peek(struct FFBufQueue *queue, unsigned index)
Get a buffer from the queue without altering it.
#define AV_QSORT(p, num, type, cmp)
Quicksort This sort is fast, and fully inplace but not stable and it is possible to construct input t...
static av_always_inline av_const double round(double x)
Structure holding the queue.
int w
agreed upon image width
it s the only field you need to keep assuming you have a context There is some magic you don t need to care about around this just let it vf default value
static av_always_inline float cbrtf(float x)
@ AV_PIX_FMT_YUVJ440P
planar YUV 4:4:0 full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV440P and setting color_range
const char * name
Pad name.
void * av_calloc(size_t nmemb, size_t size)
#define AV_PIX_FMT_YUV444P9
static enum AVPixelFormat pixel_fmts[]
int(* deflicker)(AVFilterContext *ctx, const uint8_t *src, ptrdiff_t src_linesize, uint8_t *dst, ptrdiff_t dst_linesize, int w, int h, float f)
#define AV_PIX_FMT_YUVA444P9
#define AV_PIX_FMT_YUV420P12
#define AV_PIX_FMT_YUV422P14
int h
agreed upon image height
#define AV_PIX_FMT_YUVA422P12
static void get_qm_factor(AVFilterContext *ctx, float *f)
@ AV_PIX_FMT_YUV444P
planar YUV 4:4:4, 24bpp, (1 Cr & Cb sample per 1x1 Y samples)
@ AV_PIX_FMT_YUV422P
planar YUV 4:2:2, 16bpp, (1 Cr & Cb sample per 2x1 Y samples)
Descriptor that unambiguously describes how the bits of a pixel are stored in the up to 4 data planes...
static void get_cm_factor(AVFilterContext *ctx, float *f)
#define FILTER_OUTPUTS(array)
int av_dict_set(AVDictionary **pm, const char *key, const char *value, int flags)
Set the given entry in *pm, overwriting an existing entry.
@ AV_PIX_FMT_YUV411P
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples)
static int config_input(AVFilterLink *inlink)
int linesize[AV_NUM_DATA_POINTERS]
For video, a positive or negative value, which is typically indicating the size in bytes of each pict...
@ AV_PIX_FMT_YUV410P
planar YUV 4:1:0, 9bpp, (1 Cr & Cb sample per 4x4 Y samples)
#define AV_PIX_FMT_YUV440P12
#define AV_PIX_FMT_YUV444P14
#define AV_PIX_FMT_GRAY12
static float calc_avgy16(AVFilterContext *ctx, AVFrame *in)
@ AV_PIX_FMT_YUVA422P
planar YUV 4:2:2 24bpp, (1 Cr & Cb sample per 2x1 Y & A samples)
static void get_hm_factor(AVFilterContext *ctx, float *f)
#define AV_PIX_FMT_YUV420P14
Generated on Wed Aug 24 2022 21:42:04 for FFmpeg by
doxygen
1.8.17