Go to the documentation of this file.
38 #define OFFSET(x) offsetof(SignatureContext, x)
39 #define FLAGS AV_OPT_FLAG_FILTERING_PARAM | AV_OPT_FLAG_VIDEO_PARAM
40 #define BLOCK_LCM (int64_t) 476985600
43 {
"detectmode",
"set the detectmode",
48 {
"nb_inputs",
"number of inputs",
50 {
"filename",
"filename for output files",
52 {
"format",
"set output format",
56 {
"th_d",
"threshold to detect one word as similar",
58 {
"th_dc",
"threshold to detect all words as similar",
60 {
"th_xh",
"threshold to detect frames as similar",
62 {
"th_di",
"minimum length of matching sequence in frames",
64 {
"th_it",
"threshold for relation of good to all frames",
103 return (
b->to.y -
b->up.y + 1) * (
b->to.x -
b->up.x + 1);
117 if (x0-1 >= 0 && y0-1 >= 0) {
118 sum = intpic[y1][x1] + intpic[y0-1][x0-1] - intpic[y1][x0-1] - intpic[y0-1][x1];
119 }
else if (x0-1 >= 0) {
120 sum = intpic[y1][x1] - intpic[y1][x0-1];
121 }
else if (y0-1 >= 0) {
122 sum = intpic[y1][x1] - intpic[y0-1][x1];
124 sum = intpic[y1][x1];
129 static int cmp(
const void *x,
const void *y)
131 const uint64_t *
a = x, *
b = y;
132 return *
a < *
b ? -1 : ( *
a > *
b ? 1 : 0 );
151 static const uint8_t pot3[5] = { 3*3*3*3, 3*3*3, 3*3, 3, 1 };
155 static const unsigned int wordvec[25] = {44,57,70,100,101,102,103,111,175,210,217,219,233,237,269,270,273,274,275,285,295,296,334,337,354};
156 static const uint8_t s2usw[25] = { 5,10,11, 15, 20, 21, 12, 22, 6, 0, 1, 2, 7, 13, 14, 8, 9, 3, 23, 16, 17, 24, 4, 18, 19};
158 uint8_t wordt2b[5] = { 0, 0, 0, 0, 0 };
159 uint64_t intpic[32][32];
161 uint8_t *p = picref->
data[0];
166 int f = 0,
g = 0,
w = 0;
167 int32_t dh1 = 1, dh2 = 1, dw1 = 1, dw2 = 1,
a,
b;
169 int i, j, k, ternary;
190 fs->pts = picref->
pts;
193 memset(intpic, 0,
sizeof(uint64_t)*32*32);
205 intpic[inti][intj] += p[j];
223 for (
i = 0;
i < 32;
i++) {
229 a = (
a == dh1)? dh2 : dh1;
231 for (j = 0; j < 32; j++) {
236 b = (
b == dw1)? dw2 : dw1;
238 rowcount += intpic[
i][j] *
a *
b * precfactor / denom;
240 intpic[
i][j] = intpic[
i-1][j] + rowcount;
242 intpic[
i][j] = rowcount;
252 uint64_t* sortsignature;
257 sortsignature = elemsignature + elemcat->
elem_count;
266 sum = blocksum / blocksize;
268 sum -= 128 * precfactor * denom;
276 sum -= blocksum / blocksize;
277 conflist[
g++] =
FFABS(sum * 8 / (precfactor * denom));
280 elemsignature[j] = sum;
281 sortsignature[j] =
FFABS(sum);
285 qsort(sortsignature, elemcat->
elem_count,
sizeof(uint64_t),
cmp);
290 if (elemsignature[j] < -th) {
292 }
else if (elemsignature[j] <= th) {
297 fs->framesig[
f/5] += ternary * pot3[
f%5];
299 if (
f == wordvec[
w]) {
300 fs->words[s2usw[
w]/5] += ternary * pot3[wordt2b[s2usw[
w]/5]++];
333 for (
i = 0;
i < 5;
i++) {
339 for (
i = 0;
i < 5;
i++) {
352 for (
i = 0;
i < 5;
i++) {
355 for (j = 1; j < 5; j++)
364 for (j = 1; j < 5; j++)
381 unsigned int pot3[5] = { 3*3*3*3, 3*3*3, 3*3, 3, 1 };
396 fprintf(
f,
"<?xml version='1.0' encoding='ASCII' ?>\n");
397 fprintf(
f,
"<Mpeg7 xmlns=\"urn:mpeg:mpeg7:schema:2001\" xmlns:xsi=\"http://www.w3.org/2001/XMLSchema-instance\" xsi:schemaLocation=\"urn:mpeg:mpeg7:schema:2001 schema/Mpeg7-2001.xsd\">\n");
398 fprintf(
f,
" <DescriptionUnit xsi:type=\"DescriptorCollectionType\">\n");
399 fprintf(
f,
" <Descriptor xsi:type=\"VideoSignatureType\">\n");
400 fprintf(
f,
" <VideoSignatureRegion>\n");
401 fprintf(
f,
" <VideoSignatureSpatialRegion>\n");
402 fprintf(
f,
" <Pixel>0 0 </Pixel>\n");
403 fprintf(
f,
" <Pixel>%d %d </Pixel>\n", sc->
w - 1, sc->
h - 1);
404 fprintf(
f,
" </VideoSignatureSpatialRegion>\n");
405 fprintf(
f,
" <StartFrameOfSpatialRegion>0</StartFrameOfSpatialRegion>\n");
408 fprintf(
f,
" <MediaTimeOfSpatialRegion>\n");
409 fprintf(
f,
" <StartMediaTimeOfSpatialRegion>0</StartMediaTimeOfSpatialRegion>\n");
410 fprintf(
f,
" <EndMediaTimeOfSpatialRegion>%" PRIu64
"</EndMediaTimeOfSpatialRegion>\n", sc->
coarseend->
last->
pts);
411 fprintf(
f,
" </MediaTimeOfSpatialRegion>\n");
415 fprintf(
f,
" <VSVideoSegment>\n");
416 fprintf(
f,
" <StartFrameOfSegment>%" PRIu32
"</StartFrameOfSegment>\n", cs->
first->
index);
417 fprintf(
f,
" <EndFrameOfSegment>%" PRIu32
"</EndFrameOfSegment>\n", cs->
last->
index);
418 fprintf(
f,
" <MediaTimeOfSegment>\n");
419 fprintf(
f,
" <StartMediaTimeOfSegment>%" PRIu64
"</StartMediaTimeOfSegment>\n", cs->
first->
pts);
420 fprintf(
f,
" <EndMediaTimeOfSegment>%" PRIu64
"</EndMediaTimeOfSegment>\n", cs->
last->
pts);
421 fprintf(
f,
" </MediaTimeOfSegment>\n");
422 for (
i = 0;
i < 5;
i++) {
423 fprintf(
f,
" <BagOfWords>");
424 for (j = 0; j < 31; j++) {
425 uint8_t n = cs->
data[
i][j];
427 fprintf(
f,
"%d %d %d %d %d %d %d %d ", (n & 0x80) >> 7,
437 fprintf(
f,
"%d %d %d ", (n & 0x80) >> 7,
442 fprintf(
f,
"</BagOfWords>\n");
444 fprintf(
f,
" </VSVideoSegment>\n");
449 fprintf(
f,
" <VideoFrame>\n");
450 fprintf(
f,
" <MediaTimeOfFrame>%" PRIu64
"</MediaTimeOfFrame>\n",
fs->pts);
452 fprintf(
f,
" <FrameConfidence>%d</FrameConfidence>\n",
fs->confidence);
454 fprintf(
f,
" <Word>");
455 for (
i = 0;
i < 5;
i++) {
456 fprintf(
f,
"%d ",
fs->words[
i]);
461 fprintf(
f,
"</Word>\n");
463 fprintf(
f,
" <FrameSignature>");
468 fprintf(
f,
"%d ",
fs->framesig[
i] / pot3[0]);
469 for (j = 1; j < 5; j++)
470 fprintf(
f,
" %d ",
fs->framesig[
i] % pot3[j-1] / pot3[j] );
472 fprintf(
f,
"</FrameSignature>\n");
473 fprintf(
f,
" </VideoFrame>\n");
475 fprintf(
f,
" </VideoSignatureRegion>\n");
476 fprintf(
f,
" </Descriptor>\n");
477 fprintf(
f,
" </DescriptionUnit>\n");
478 fprintf(
f,
"</Mpeg7>\n");
489 uint32_t numofsegments = (sc->
lastindex + 44)/45;
493 int len = (512 + 6 * 32 + 3*16 + 2 +
494 numofsegments * (4*32 + 1 + 5*243) +
495 sc->
lastindex * (2 + 32 + 6*8 + 608)) / 8;
532 for (
i = 0;
i < 5;
i++) {
534 for (j = 0; j < 30; j++) {
546 for (
i = 0;
i < 5;
i++) {
620 if (match.
score != 0) {
681 av_log(
ctx,
AV_LOG_ERROR,
"The filename must contain %%d or %%0nd, if you have more than one input.\n");
709 finsig = finsig->
next;
716 cousig = cousig->
next;
754 .priv_class = &signature_class,
#define AV_LOG_WARNING
Something somehow does not look correct.
AVPixelFormat
Pixel format.
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 av_unused put_bits32(PutBitContext *s, uint32_t value)
Write exactly 32 bits into a bitstream.
CoarseSignature * coarsesiglist
#define FILTER_PIXFMTS_ARRAY(array)
static const ElemCat * elements[ELEMENT_COUNT]
static int put_bytes_output(const PutBitContext *s)
int ff_filter_frame(AVFilterLink *link, AVFrame *frame)
Send a frame of data to the next filter.
#define AVERROR_EOF
End of file.
static const AVFilterPad signature_outputs[]
struct FineSignature * next
static void set_bit(uint8_t *data, size_t pos)
sets the bit at position pos to 1 in data
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
char * av_asprintf(const char *fmt,...)
static void init_put_bits(PutBitContext *s, uint8_t *buffer, int buffer_size)
Initialize the PutBitContext s.
CoarseSignature * curcoarsesig2
This structure describes decoded (raw) audio or video data.
static void put_bits(Jpeg2000EncoderContext *s, int val, int n)
put n times val bit
int64_t pts
Presentation timestamp in time_base units (time when frame should be shown to user).
int ff_request_frame(AVFilterLink *link)
Request an input frame from the filter at the other end of the link.
@ 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.
int av_get_frame_filename(char *buf, int buf_size, const char *path, int number)
uint8_t * data[AV_NUM_DATA_POINTERS]
pointer to the picture/channel planes.
Link properties exposed to filter code, but not external callers.
static int filter_frame(AVFilterLink *inlink, AVFrame *picref)
static const AVOption signature_options[]
int av_strerror(int errnum, char *errbuf, size_t errbuf_size)
Put a description of the AVERROR code errnum in errbuf.
static MatchingInfo lookup_signatures(AVFilterContext *ctx, SignatureContext *sc, StreamContext *first, StreamContext *second, int mode)
#define AVFILTER_FLAG_DYNAMIC_INPUTS
The number of the filter inputs is not determined just by AVFilter.inputs.
struct FineSignature * last
A filter pad used for either input or output.
static const char signature[]
@ AV_PIX_FMT_YUVJ411P
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples) full scale (JPEG), deprecated in favor ...
#define AV_LOG_ERROR
Something went wrong and cannot losslessly be recovered.
static uint64_t get_block_sum(StreamContext *sc, uint64_t intpic[32][32], const Block *b)
@ AV_PIX_FMT_YUVJ422P
planar YUV 4:2:2, 16bpp, full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV422P and setting col...
struct FineSignature * first
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 format(the sample packing is implied by the sample format) and sample rate. The lists are not just lists
@ AV_OPT_TYPE_DOUBLE
Underlying C type is double.
AVRational sample_aspect_ratio
agreed upon sample aspect ratio
#define av_assert0(cond)
assert() equivalent, that is always enabled.
#define AV_LOG_DEBUG
Stuff which is only useful for libav* developers.
static int export(AVFilterContext *ctx, StreamContext *sc, int input)
@ AV_PIX_FMT_YUV420P
planar YUV 4:2:0, 12bpp, (1 Cr & Cb sample per 2x2 Y samples)
#define FILTER_OUTPUTS(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 FFABS(a)
Absolute value, Note, INT_MIN / INT64_MIN result in undefined behavior as they are not representable ...
const AVFilter ff_vf_signature
static int config_output(AVFilterLink *outlink)
int av_log_get_level(void)
Get the current log level.
#define fs(width, name, subs,...)
static int request_frame(AVFilterLink *outlink)
int ff_append_inpad_free_name(AVFilterContext *f, AVFilterPad *p)
@ 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 av_cold void uninit(AVFilterContext *ctx)
static int cmp(const void *x, const void *y)
@ AV_PIX_FMT_GRAY8
Y , 8bpp.
static int get_block_size(const Block *b)
AVFILTER_DEFINE_CLASS(signature)
static FilterLink * ff_filter_link(AVFilterLink *link)
#define NULL_IF_CONFIG_SMALL(x)
Return NULL if CONFIG_SMALL is true, otherwise the argument without modification.
static int config_input(AVFilterLink *inlink)
static enum AVPixelFormat pix_fmts[]
FineSignature * curfinesig
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
and forward the test the status of outputs and forward it to the corresponding return FFERROR_NOT_READY If the filters stores internally one or a few frame for some input
#define AV_LOG_INFO
Standard information.
#define i(width, name, range_min, range_max)
#define FF_INLINK_IDX(link)
Find the index of a link.
int w
agreed upon image width
#define av_malloc_array(a, b)
@ AV_PIX_FMT_YUVJ440P
planar YUV 4:4:0 full scale (JPEG), deprecated in favor of AV_PIX_FMT_YUV440P and setting color_range
void * av_mallocz(size_t size)
Allocate a memory block with alignment suitable for all memory accesses (including vectors if availab...
static av_cold int init(AVFilterContext *ctx)
@ AV_PIX_FMT_NV21
as above, but U and V bytes are swapped
const char * name
Pad name.
static int xml_export(AVFilterContext *ctx, StreamContext *sc, const char *filename)
FILE * avpriv_fopen_utf8(const char *path, const char *mode)
Open a file using a UTF-8 filename.
enum AVMediaType type
AVFilterPad type.
@ AV_PIX_FMT_NV12
planar YUV 4:2:0, 12bpp, 1 plane for Y and 1 plane for the UV components, which are interleaved (firs...
CoarseSignature * coarseend
struct CoarseSignature * next
int h
agreed upon image height
the frame and frame reference mechanism is intended to as much as expensive copies of that data while still allowing the filters to produce correct results The data is stored in buffers represented by AVFrame structures Several references can point to the same frame buffer
@ AV_OPT_TYPE_INT
Underlying C type is int.
static int binary_export(AVFilterContext *ctx, StreamContext *sc, const char *filename)
struct FineSignature * second
AVRational time_base
Define the time base used by the PTS of the frames/samples which will pass through this link.
@ AV_PIX_FMT_YUV444P
planar YUV 4:4:4, 24bpp, (1 Cr & Cb sample per 1x1 Y samples)
CoarseSignature * curcoarsesig1
@ AV_PIX_FMT_YUV422P
planar YUV 4:2:2, 16bpp, (1 Cr & Cb sample per 2x1 Y samples)
static void flush_put_bits(PutBitContext *s)
Pad the end of the output stream with zeros.
@ AV_PIX_FMT_YUV411P
planar YUV 4:1:1, 12bpp, (1 Cr & Cb sample per 4x1 Y samples)
struct FineSignature * first
FineSignature * finesiglist
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)
size_t av_strlcpy(char *dst, const char *src, size_t size)
Copy the string src to dst, but no more than size - 1 bytes, and null-terminate dst.
StreamContext * streamcontexts
AVRational frame_rate
Frame rate of the stream on the link, or 1/0 if unknown or variable.
@ AV_OPT_TYPE_STRING
Underlying C type is a uint8_t* that is either NULL or points to a C string allocated with the av_mal...
@ AV_OPT_TYPE_CONST
Special option type for declaring named constants.