/
main.cpp
1167 lines (999 loc) · 51.8 KB
/
main.cpp
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
680
681
682
683
684
685
686
687
688
689
690
691
692
693
694
695
696
697
698
699
700
701
702
703
704
705
706
707
708
709
710
711
712
713
714
715
716
717
718
719
720
721
722
723
724
725
726
727
728
729
730
731
732
733
734
735
736
737
738
739
740
741
742
743
744
745
746
747
748
749
750
751
752
753
754
755
756
757
758
759
760
761
762
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787
788
789
790
791
792
793
794
795
796
797
798
799
800
801
802
803
804
805
806
807
808
809
810
811
812
813
814
815
816
817
818
819
820
821
822
823
824
825
826
827
828
829
830
831
832
833
834
835
836
837
838
839
840
841
842
843
844
845
846
847
848
849
850
851
852
853
854
855
856
857
858
859
860
861
862
863
864
865
866
867
868
869
870
871
872
873
874
875
876
877
878
879
880
881
882
883
884
885
886
887
888
889
890
891
892
893
894
895
896
897
898
899
900
901
902
903
904
905
906
907
908
909
910
911
912
913
914
915
916
917
918
919
920
921
922
923
924
925
926
927
928
929
930
931
932
933
934
935
936
937
938
939
940
941
942
943
944
945
946
947
948
949
950
951
952
953
954
955
956
957
958
959
960
961
962
963
964
965
966
967
968
969
970
971
972
973
974
975
976
977
978
979
980
981
982
983
984
985
986
987
988
989
990
991
992
993
994
995
996
997
998
999
1000
#include <opencv2/core.hpp>
#include <opencv2/imgcodecs.hpp>
#include <opencv2/highgui.hpp>
#include <opencv2/xfeatures2d/nonfree.hpp>
#include <iostream>
#include "GaussianBlur_modified.hpp"
#include "substraction_kernel.hpp"
#include "helper.hpp"
using namespace cv;
using namespace std;
#include "precomp.hpp"
#include <stdarg.h>
#include <opencv2/core/hal/hal.hpp>
// define global variables
int timer_helper_subpixel = 10;
uint64_t cycles_subpixel = 0;
int timer_helper_harris = 10;
uint64_t cycles_harris = 0;
long long Extrema_Localization_count = 0;
long long Adjusting_keypoint_locations_count = 0;
long long Eliminating_Edge_Responses_count = 0;
uint64_t cycles_conv = 0;
uint64_t cycles_mem = 0;
uint64_t buildGaussianPyramidCycles = 0;
uint64_t buildDoGCycles = 0;
// source code of SIFT from opencv
namespace cv
{
namespace xfeatures2d
{
/*!
SIFT implementation.
The class implements SIFT algorithm by D. Lowe.
*/
class SIFT_Impl : public SIFT
{
public:
explicit SIFT_Impl( int nfeatures = 0, int nOctaveLayers = 3,
double contrastThreshold = 0.04, double edgeThreshold = 10,
double sigma = 1.6);
//! returns the descriptor size in floats (128)
int descriptorSize() const CV_OVERRIDE;
//! returns the descriptor type
int descriptorType() const CV_OVERRIDE;
//! returns the default norm type
int defaultNorm() const CV_OVERRIDE;
//! finds the keypoints and computes descriptors for them using SIFT algorithm.
//! Optionally it can compute descriptors for the user-provided keypoints
void detectAndCompute(InputArray img, InputArray mask,
std::vector<KeyPoint>& keypoints,
OutputArray descriptors,
bool useProvidedKeypoints = false) CV_OVERRIDE;
void buildGaussianPyramid( const Mat& base, std::vector<Mat>& pyr, int nOctaves ) const;
void buildDoGPyramid( const std::vector<Mat>& pyr, std::vector<Mat>& dogpyr ) const;
void findScaleSpaceExtrema( const std::vector<Mat>& gauss_pyr, const std::vector<Mat>& dog_pyr,
std::vector<KeyPoint>& keypoints ) const;
protected:
CV_PROP_RW int nfeatures;
CV_PROP_RW int nOctaveLayers;
CV_PROP_RW double contrastThreshold;
CV_PROP_RW double edgeThreshold;
CV_PROP_RW double sigma;
};
Ptr<SIFT> SIFT::create( int _nfeatures, int _nOctaveLayers,
double _contrastThreshold, double _edgeThreshold, double _sigma )
{
return makePtr<SIFT_Impl>(_nfeatures, _nOctaveLayers, _contrastThreshold, _edgeThreshold, _sigma);
}
/******************************* Defs and macros *****************************/
// default width of descriptor histogram array
static const int SIFT_DESCR_WIDTH = 4;
// default number of bins per histogram in descriptor array
static const int SIFT_DESCR_HIST_BINS = 8;
// assumed gaussian blur for input image
static const float SIFT_INIT_SIGMA = 0.5f;
// width of border in which to ignore keypoints
static const int SIFT_IMG_BORDER = 5;
// maximum steps of keypoint interpolation before failure
static const int SIFT_MAX_INTERP_STEPS = 5;
// default number of bins in histogram for orientation assignment
static const int SIFT_ORI_HIST_BINS = 36;
// determines gaussian sigma for orientation assignment
static const float SIFT_ORI_SIG_FCTR = 1.5f;
// determines the radius of the region used in orientation assignment
static const float SIFT_ORI_RADIUS = 3 * SIFT_ORI_SIG_FCTR;
// orientation magnitude relative to max that results in new feature
static const float SIFT_ORI_PEAK_RATIO = 0.8f;
// determines the size of a single descriptor orientation histogram
static const float SIFT_DESCR_SCL_FCTR = 3.f;
// threshold on magnitude of elements of descriptor vector
static const float SIFT_DESCR_MAG_THR = 0.2f;
// factor used to convert floating-point descriptor to unsigned char
static const float SIFT_INT_DESCR_FCTR = 512.f;
#define DoG_TYPE_SHORT 0
#if DoG_TYPE_SHORT
// intermediate type used for DoG pyramids
typedef short sift_wt;
static const int SIFT_FIXPT_SCALE = 48;
#else
// intermediate type used for DoG pyramids
typedef float sift_wt;
static const int SIFT_FIXPT_SCALE = 1;
#endif
static inline void
unpackOctave(const KeyPoint& kpt, int& octave, int& layer, float& scale)
{
octave = kpt.octave & 255;
layer = (kpt.octave >> 8) & 255;
octave = octave < 128 ? octave : (-128 | octave);
scale = octave >= 0 ? 1.f/(1 << octave) : (float)(1 << -octave);
}
static Mat createInitialImage( const Mat& img, bool doubleImageSize, float sigma )
{
Mat gray, gray_fpt;
if( img.channels() == 3 || img.channels() == 4 )
{
cvtColor(img, gray, COLOR_BGR2GRAY);
gray.convertTo(gray_fpt, DataType<sift_wt>::type, SIFT_FIXPT_SCALE, 0);
}
else
img.convertTo(gray_fpt, DataType<sift_wt>::type, SIFT_FIXPT_SCALE, 0);
float sig_diff;
if( doubleImageSize )
{
sig_diff = sqrtf( std::max(sigma * sigma - SIFT_INIT_SIGMA * SIFT_INIT_SIGMA * 4, 0.01f) );
Mat dbl;
#if DoG_TYPE_SHORT
resize(gray_fpt, dbl, Size(gray_fpt.cols*2, gray_fpt.rows*2), 0, 0, INTER_LINEAR_EXACT);
#else
resize(gray_fpt, dbl, Size(gray_fpt.cols*2, gray_fpt.rows*2), 0, 0, INTER_LINEAR);
#endif
GaussianBlur(dbl, dbl, Size(), sig_diff, sig_diff);
return dbl;
}
else
{
sig_diff = sqrtf( std::max(sigma * sigma - SIFT_INIT_SIGMA * SIFT_INIT_SIGMA, 0.01f) );
GaussianBlur(gray_fpt, gray_fpt, Size(), sig_diff, sig_diff);
return gray_fpt;
}
}
void SIFT_Impl::buildGaussianPyramid( const Mat& base, std::vector<Mat>& pyr, int nOctaves ) const
{
std::vector<double> sig(nOctaveLayers + 3);
pyr.resize(nOctaves*(nOctaveLayers + 3));
// precompute Gaussian sigmas using the following formula:
// \sigma_{total}^2 = \sigma_{i}^2 + \sigma_{i-1}^2
sig[0] = sigma;
double k = std::pow( 2., 1. / nOctaveLayers );
for( int i = 1; i < nOctaveLayers + 3; i++ )
{
double sig_prev = std::pow(k, (double)(i-1))*sigma;
double sig_total = sig_prev*k;
sig[i] = std::sqrt(sig_total*sig_total - sig_prev*sig_prev);
}
for( int o = 0; o < nOctaves; o++ )
{
for( int i = 0; i < nOctaveLayers + 3; i++ )
{
Mat& dst = pyr[o*(nOctaveLayers + 3) + i];
if( o == 0 && i == 0 ) {
dst = base;
}
// base of new octave is halved image from end of previous octave
else if( i == 0 )
{
const Mat& src = pyr[(o-1)*(nOctaveLayers + 3) + nOctaveLayers];
resize(src, dst, Size(src.cols/2, src.rows/2),
0, 0, INTER_NEAREST); // resize image when changing to another octave
}
else
{
const Mat& src = pyr[o*(nOctaveLayers + 3) + i-1];
// GaussianBlur(src, dst, Size(), sig[i], sig[i]);
GaussianBlur_modified(src, dst, Size(), sig[i], sig[i]);
}
}
}
}
class buildDoGPyramidComputer : public ParallelLoopBody
{
public:
buildDoGPyramidComputer(
int _nOctaveLayers,
const std::vector<Mat>& _gpyr,
std::vector<Mat>& _dogpyr)
: nOctaveLayers(_nOctaveLayers),
gpyr(_gpyr),
dogpyr(_dogpyr) { }
void operator()( const cv::Range& range ) const CV_OVERRIDE
{
const int begin = range.start;
const int end = range.end;
for( int a = begin; a < end; a++ )
{
const int o = a / (nOctaveLayers + 2); // o-th octave
const int i = a % (nOctaveLayers + 2); // i-th layer
const Mat& src1 = gpyr[o*(nOctaveLayers + 3) + i];
const Mat& src2 = gpyr[o*(nOctaveLayers + 3) + i + 1];
Mat& dst = dogpyr[o*(nOctaveLayers + 2) + i];
// subtract(src2, src1, dst, noArray(), DataType<sift_wt>::type);
substraction_kernel(src1, src2, dst);
}
}
private:
int nOctaveLayers;
const std::vector<Mat>& gpyr;
std::vector<Mat>& dogpyr;
};
void SIFT_Impl::buildDoGPyramid( const std::vector<Mat>& gpyr, std::vector<Mat>& dogpyr ) const
{
int nOctaves = (int)gpyr.size()/(nOctaveLayers + 3);
dogpyr.resize( nOctaves*(nOctaveLayers + 2) );
parallel_for_(Range(0, nOctaves * (nOctaveLayers + 2)), buildDoGPyramidComputer(nOctaveLayers, gpyr, dogpyr));
// range is DOG pyramid size
}
// Computes a gradient orientation histogram at a specified pixel
static float calcOrientationHist( const Mat& img, Point pt, int radius,
float sigma, float* hist, int n )
{
int i, j, k, len = (radius*2+1)*(radius*2+1);
float expf_scale = -1.f/(2.f * sigma * sigma);
AutoBuffer<float> buf(len*4 + n+4);
float *X = buf.data(), *Y = X + len, *Mag = X, *Ori = Y + len, *W = Ori + len;
float* temphist = W + len + 2;
for( i = 0; i < n; i++ )
temphist[i] = 0.f;
for( i = -radius, k = 0; i <= radius; i++ )
{
int y = pt.y + i;
if( y <= 0 || y >= img.rows - 1 )
continue;
for( j = -radius; j <= radius; j++ )
{
int x = pt.x + j;
if( x <= 0 || x >= img.cols - 1 )
continue;
float dx = (float)(img.at<sift_wt>(y, x+1) - img.at<sift_wt>(y, x-1));
float dy = (float)(img.at<sift_wt>(y-1, x) - img.at<sift_wt>(y+1, x));
X[k] = dx; Y[k] = dy; W[k] = (i*i + j*j)*expf_scale;
k++;
}
}
len = k;
// compute gradient values, orientations and the weights over the pixel neighborhood
cv::hal::exp32f(W, W, len);
cv::hal::fastAtan2(Y, X, Ori, len, true);
cv::hal::magnitude32f(X, Y, Mag, len);
k = 0;
#if CV_AVX2
if( USE_AVX2 )
{
__m256 __nd360 = _mm256_set1_ps(n/360.f);
__m256i __n = _mm256_set1_epi32(n);
int CV_DECL_ALIGNED(32) bin_buf[8];
float CV_DECL_ALIGNED(32) w_mul_mag_buf[8];
for ( ; k <= len - 8; k+=8 )
{
__m256i __bin = _mm256_cvtps_epi32(_mm256_mul_ps(__nd360, _mm256_loadu_ps(&Ori[k])));
__bin = _mm256_sub_epi32(__bin, _mm256_andnot_si256(_mm256_cmpgt_epi32(__n, __bin), __n));
__bin = _mm256_add_epi32(__bin, _mm256_and_si256(__n, _mm256_cmpgt_epi32(_mm256_setzero_si256(), __bin)));
__m256 __w_mul_mag = _mm256_mul_ps(_mm256_loadu_ps(&W[k]), _mm256_loadu_ps(&Mag[k]));
_mm256_store_si256((__m256i *) bin_buf, __bin);
_mm256_store_ps(w_mul_mag_buf, __w_mul_mag);
temphist[bin_buf[0]] += w_mul_mag_buf[0];
temphist[bin_buf[1]] += w_mul_mag_buf[1];
temphist[bin_buf[2]] += w_mul_mag_buf[2];
temphist[bin_buf[3]] += w_mul_mag_buf[3];
temphist[bin_buf[4]] += w_mul_mag_buf[4];
temphist[bin_buf[5]] += w_mul_mag_buf[5];
temphist[bin_buf[6]] += w_mul_mag_buf[6];
temphist[bin_buf[7]] += w_mul_mag_buf[7];
}
}
#endif
for( ; k < len; k++ )
{
int bin = cvRound((n/360.f)*Ori[k]);
if( bin >= n )
bin -= n;
if( bin < 0 )
bin += n;
temphist[bin] += W[k]*Mag[k];
}
// smooth the histogram
temphist[-1] = temphist[n-1];
temphist[-2] = temphist[n-2];
temphist[n] = temphist[0];
temphist[n+1] = temphist[1];
i = 0;
#if CV_AVX2
if( USE_AVX2 )
{
__m256 __d_1_16 = _mm256_set1_ps(1.f/16.f);
__m256 __d_4_16 = _mm256_set1_ps(4.f/16.f);
__m256 __d_6_16 = _mm256_set1_ps(6.f/16.f);
for( ; i <= n - 8; i+=8 )
{
#if CV_FMA3
__m256 __hist = _mm256_fmadd_ps(
_mm256_add_ps(_mm256_loadu_ps(&temphist[i-2]), _mm256_loadu_ps(&temphist[i+2])),
__d_1_16,
_mm256_fmadd_ps(
_mm256_add_ps(_mm256_loadu_ps(&temphist[i-1]), _mm256_loadu_ps(&temphist[i+1])),
__d_4_16,
_mm256_mul_ps(_mm256_loadu_ps(&temphist[i]), __d_6_16)));
#else
__m256 __hist = _mm256_add_ps(
_mm256_mul_ps(
_mm256_add_ps(_mm256_loadu_ps(&temphist[i-2]), _mm256_loadu_ps(&temphist[i+2])),
__d_1_16),
_mm256_add_ps(
_mm256_mul_ps(
_mm256_add_ps(_mm256_loadu_ps(&temphist[i-1]), _mm256_loadu_ps(&temphist[i+1])),
__d_4_16),
_mm256_mul_ps(_mm256_loadu_ps(&temphist[i]), __d_6_16)));
#endif
_mm256_storeu_ps(&hist[i], __hist);
}
}
#endif
for( ; i < n; i++ )
{
hist[i] = (temphist[i-2] + temphist[i+2])*(1.f/16.f) +
(temphist[i-1] + temphist[i+1])*(4.f/16.f) +
temphist[i]*(6.f/16.f);
}
float maxval = hist[0];
for( i = 1; i < n; i++ )
maxval = std::max(maxval, hist[i]);
return maxval;
}
//
// Interpolates a scale-space extremum's location and scale to subpixel
// accuracy to form an image feature. Rejects features with low contrast.
// Based on Section 4 of Lowe's paper.
static bool adjustLocalExtrema( const std::vector<Mat>& dog_pyr, KeyPoint& kpt, int octv,
int& layer, int& r, int& c, int nOctaveLayers,
float contrastThreshold, float edgeThreshold, float sigma )
{
const float img_scale = 1.f/(255*SIFT_FIXPT_SCALE);
const float deriv_scale = img_scale*0.5f;
const float second_deriv_scale = img_scale;
const float cross_deriv_scale = img_scale*0.25f;
float xi=0, xr=0, xc=0, contr=0;
int i = 0;
uint64_t start, end;
for( ; i < SIFT_MAX_INTERP_STEPS; i++ )
{
Adjusting_keypoint_locations_count ++;
start = rdtsc();
int idx = octv*(nOctaveLayers+2) + layer;
const Mat& img = dog_pyr[idx];
const Mat& prev = dog_pyr[idx-1];
const Mat& next = dog_pyr[idx+1];
Vec3f dD((img.at<sift_wt>(r, c+1) - img.at<sift_wt>(r, c-1))*deriv_scale,
(img.at<sift_wt>(r+1, c) - img.at<sift_wt>(r-1, c))*deriv_scale,
(next.at<sift_wt>(r, c) - prev.at<sift_wt>(r, c))*deriv_scale);
// first order derivative
float v2 = (float)img.at<sift_wt>(r, c)*2;
float dxx = (img.at<sift_wt>(r, c+1) + img.at<sift_wt>(r, c-1) - v2)*second_deriv_scale;
float dyy = (img.at<sift_wt>(r+1, c) + img.at<sift_wt>(r-1, c) - v2)*second_deriv_scale;
float dss = (next.at<sift_wt>(r, c) + prev.at<sift_wt>(r, c) - v2)*second_deriv_scale;
float dxy = (img.at<sift_wt>(r+1, c+1) - img.at<sift_wt>(r+1, c-1) -
img.at<sift_wt>(r-1, c+1) + img.at<sift_wt>(r-1, c-1))*cross_deriv_scale;
float dxs = (next.at<sift_wt>(r, c+1) - next.at<sift_wt>(r, c-1) -
prev.at<sift_wt>(r, c+1) + prev.at<sift_wt>(r, c-1))*cross_deriv_scale;
float dys = (next.at<sift_wt>(r+1, c) - next.at<sift_wt>(r-1, c) -
prev.at<sift_wt>(r+1, c) + prev.at<sift_wt>(r-1, c))*cross_deriv_scale;
Matx33f H(dxx, dxy, dxs,
dxy, dyy, dys,
dxs, dys, dss);
Vec3f X = H.solve(dD, DECOMP_LU);
xi = -X[2];
xr = -X[1];
xc = -X[0];
end = rdtsc();
if (timer_helper_subpixel-- > 0) {
cycles_subpixel += (end-start);
if(timer_helper_subpixel == 0){
}
}
if( std::abs(xi) < 0.5f && std::abs(xr) < 0.5f && std::abs(xc) < 0.5f )
break;
if( std::abs(xi) > (float)(INT_MAX/3) ||
std::abs(xr) > (float)(INT_MAX/3) ||
std::abs(xc) > (float)(INT_MAX/3) )
return false;
c += cvRound(xc);
r += cvRound(xr);
layer += cvRound(xi);
if( layer < 1 || layer > nOctaveLayers ||
c < SIFT_IMG_BORDER || c >= img.cols - SIFT_IMG_BORDER ||
r < SIFT_IMG_BORDER || r >= img.rows - SIFT_IMG_BORDER )
return false;
}
// ensure convergence of interpolation
if( i >= SIFT_MAX_INTERP_STEPS )
return false;
{
Eliminating_Edge_Responses_count ++;
start = rdtsc();
int idx = octv*(nOctaveLayers+2) + layer; //2 add 1mul
const Mat& img = dog_pyr[idx];
const Mat& prev = dog_pyr[idx-1]; // 1add
const Mat& next = dog_pyr[idx+1]; // 1add
Matx31f dD((img.at<sift_wt>(r, c+1) - img.at<sift_wt>(r, c-1))*deriv_scale,
(img.at<sift_wt>(r+1, c) - img.at<sift_wt>(r-1, c))*deriv_scale,
(next.at<sift_wt>(r, c) - prev.at<sift_wt>(r, c))*deriv_scale);
float t = dD.dot(Matx31f(xc, xr, xi));
contr = img.at<sift_wt>(r, c)*img_scale + t * 0.5f; //1add 2mul
if( std::abs( contr ) * nOctaveLayers < contrastThreshold ) //1add 1mul
return false;
// principal curvatures are computed using the trace and det of Hessian
float v2 = img.at<sift_wt>(r, c)*2.f;
float dxx = (img.at<sift_wt>(r, c+1) + img.at<sift_wt>(r, c-1) - v2)*second_deriv_scale;
float dyy = (img.at<sift_wt>(r+1, c) + img.at<sift_wt>(r-1, c) - v2)*second_deriv_scale;
float dxy = (img.at<sift_wt>(r+1, c+1) - img.at<sift_wt>(r+1, c-1) -
img.at<sift_wt>(r-1, c+1) + img.at<sift_wt>(r-1, c-1)) * cross_deriv_scale;
float tr = dxx + dyy;
float det = dxx * dyy - dxy * dxy;
end = rdtsc();
if (timer_helper_harris-- > 0) {
cycles_harris += (end-start);
if(timer_helper_harris == 0){
}
}
if( det <= 0 || tr*tr*edgeThreshold >= (edgeThreshold + 1)*(edgeThreshold + 1)*det )
return false;
}
kpt.pt.x = (c + xc) * (1 << octv);
kpt.pt.y = (r + xr) * (1 << octv);
kpt.octave = octv + (layer << 8) + (cvRound((xi + 0.5)*255) << 16);
kpt.size = sigma*powf(2.f, (layer + xi) / nOctaveLayers)*(1 << octv)*2;
kpt.response = std::abs(contr);
return true;
}
class findScaleSpaceExtremaComputer : public ParallelLoopBody
{
public:
findScaleSpaceExtremaComputer(
int _o,
int _i,
int _threshold,
int _idx,
int _step,
int _cols,
int _nOctaveLayers,
double _contrastThreshold,
double _edgeThreshold,
double _sigma,
const std::vector<Mat>& _gauss_pyr,
const std::vector<Mat>& _dog_pyr,
TLSData<std::vector<KeyPoint> > &_tls_kpts_struct)
: o(_o),
i(_i),
threshold(_threshold),
idx(_idx),
step(_step),
cols(_cols),
nOctaveLayers(_nOctaveLayers),
contrastThreshold(_contrastThreshold),
edgeThreshold(_edgeThreshold),
sigma(_sigma),
gauss_pyr(_gauss_pyr),
dog_pyr(_dog_pyr),
tls_kpts_struct(_tls_kpts_struct) { }
void operator()( const cv::Range& range ) const CV_OVERRIDE
{
const int begin = range.start;
const int end = range.end;
static const int n = SIFT_ORI_HIST_BINS;
float hist[n];
const Mat& img = dog_pyr[idx];
const Mat& prev = dog_pyr[idx-1];
const Mat& next = dog_pyr[idx+1];
std::vector<KeyPoint> *tls_kpts = tls_kpts_struct.get();
KeyPoint kpt;
for( int r = begin; r < end; r++)
{
const sift_wt* currptr = img.ptr<sift_wt>(r);
const sift_wt* prevptr = prev.ptr<sift_wt>(r);
const sift_wt* nextptr = next.ptr<sift_wt>(r);
for( int c = SIFT_IMG_BORDER; c < cols-SIFT_IMG_BORDER; c++)
{
sift_wt val = currptr[c];
Extrema_Localization_count ++;
// find local extrema with pixel accuracy
if( std::abs(val) > threshold &&
((val > 0 && val >= currptr[c-1] && val >= currptr[c+1] &&
val >= currptr[c-step-1] && val >= currptr[c-step] && val >= currptr[c-step+1] &&
val >= currptr[c+step-1] && val >= currptr[c+step] && val >= currptr[c+step+1] &&
val >= nextptr[c] && val >= nextptr[c-1] && val >= nextptr[c+1] &&
val >= nextptr[c-step-1] && val >= nextptr[c-step] && val >= nextptr[c-step+1] &&
val >= nextptr[c+step-1] && val >= nextptr[c+step] && val >= nextptr[c+step+1] &&
val >= prevptr[c] && val >= prevptr[c-1] && val >= prevptr[c+1] &&
val >= prevptr[c-step-1] && val >= prevptr[c-step] && val >= prevptr[c-step+1] &&
val >= prevptr[c+step-1] && val >= prevptr[c+step] && val >= prevptr[c+step+1]) ||
(val < 0 && val <= currptr[c-1] && val <= currptr[c+1] &&
val <= currptr[c-step-1] && val <= currptr[c-step] && val <= currptr[c-step+1] &&
val <= currptr[c+step-1] && val <= currptr[c+step] && val <= currptr[c+step+1] &&
val <= nextptr[c] && val <= nextptr[c-1] && val <= nextptr[c+1] &&
val <= nextptr[c-step-1] && val <= nextptr[c-step] && val <= nextptr[c-step+1] &&
val <= nextptr[c+step-1] && val <= nextptr[c+step] && val <= nextptr[c+step+1] &&
val <= prevptr[c] && val <= prevptr[c-1] && val <= prevptr[c+1] &&
val <= prevptr[c-step-1] && val <= prevptr[c-step] && val <= prevptr[c-step+1] &&
val <= prevptr[c+step-1] && val <= prevptr[c+step] && val <= prevptr[c+step+1])))
{
int r1 = r, c1 = c, layer = i;
//TODO: ask TA about how to optimize/benchmarking it?
if( !adjustLocalExtrema(dog_pyr, kpt, o, layer, r1, c1,
nOctaveLayers, (float)contrastThreshold,
(float)edgeThreshold, (float)sigma) )
continue;
float scl_octv = kpt.size*0.5f/(1 << o);
float omax = calcOrientationHist(gauss_pyr[o*(nOctaveLayers+3) + layer],
Point(c1, r1),
cvRound(SIFT_ORI_RADIUS * scl_octv),
SIFT_ORI_SIG_FCTR * scl_octv,
hist, n);
float mag_thr = (float)(omax * SIFT_ORI_PEAK_RATIO);
for( int j = 0; j < n; j++ )
{
int l = j > 0 ? j - 1 : n - 1;
int r2 = j < n-1 ? j + 1 : 0;
if( hist[j] > hist[l] && hist[j] > hist[r2] && hist[j] >= mag_thr )
{
float bin = j + 0.5f * (hist[l]-hist[r2]) / (hist[l] - 2*hist[j] + hist[r2]);
bin = bin < 0 ? n + bin : bin >= n ? bin - n : bin;
kpt.angle = 360.f - (float)((360.f/n) * bin);
if(std::abs(kpt.angle - 360.f) < FLT_EPSILON)
kpt.angle = 0.f;
{
tls_kpts->push_back(kpt);
}
}
}
}
}
}
}
private:
int o, i;
int threshold;
int idx, step, cols;
int nOctaveLayers;
double contrastThreshold;
double edgeThreshold;
double sigma;
const std::vector<Mat>& gauss_pyr;
const std::vector<Mat>& dog_pyr;
TLSData<std::vector<KeyPoint> > &tls_kpts_struct;
};
//
// Detects features at extrema in DoG scale space. Bad features are discarded
// based on contrast and ratio of principal curvatures.
void SIFT_Impl::findScaleSpaceExtrema( const std::vector<Mat>& gauss_pyr, const std::vector<Mat>& dog_pyr,
std::vector<KeyPoint>& keypoints ) const
{
const int nOctaves = (int)gauss_pyr.size()/(nOctaveLayers + 3);
const int threshold = cvFloor(0.5 * contrastThreshold / nOctaveLayers * 255 * SIFT_FIXPT_SCALE);
keypoints.clear(); // empty keypoints
TLSData<std::vector<KeyPoint> > tls_kpts_struct; // multi-threads
for( int o = 0; o < nOctaves; o++ )
for( int i = 1; i <= nOctaveLayers; i++ )
{
const int idx = o*(nOctaveLayers+2)+i;
const Mat& img = dog_pyr[idx];
const int step = (int)img.step1();
const int rows = img.rows, cols = img.cols;
parallel_for_(Range(SIFT_IMG_BORDER, rows-SIFT_IMG_BORDER),
findScaleSpaceExtremaComputer(
o, i, threshold, idx, step, cols,
nOctaveLayers,
contrastThreshold,
edgeThreshold,
sigma,
gauss_pyr, dog_pyr, tls_kpts_struct));
}
std::vector<std::vector<KeyPoint>*> kpt_vecs;
tls_kpts_struct.gather(kpt_vecs);
for (size_t i = 0; i < kpt_vecs.size(); ++i) {
keypoints.insert(keypoints.end(), kpt_vecs[i]->begin(), kpt_vecs[i]->end());
}
}
static void calcSIFTDescriptor( const Mat& img, Point2f ptf, float ori, float scl,
int d, int n, float* dst )
{
Point pt(cvRound(ptf.x), cvRound(ptf.y));
float cos_t = cosf(ori*(float)(CV_PI/180));
float sin_t = sinf(ori*(float)(CV_PI/180));
float bins_per_rad = n / 360.f;
float exp_scale = -1.f/(d * d * 0.5f);
float hist_width = SIFT_DESCR_SCL_FCTR * scl;
int radius = cvRound(hist_width * 1.4142135623730951f * (d + 1) * 0.5f);
// Clip the radius to the diagonal of the image to avoid autobuffer too large exception
radius = std::min(radius, (int) sqrt(((double) img.cols)*img.cols + ((double) img.rows)*img.rows));
cos_t /= hist_width;
sin_t /= hist_width;
int i, j, k, len = (radius*2+1)*(radius*2+1), histlen = (d+2)*(d+2)*(n+2);
int rows = img.rows, cols = img.cols;
AutoBuffer<float> buf(len*6 + histlen);
float *X = buf.data(), *Y = X + len, *Mag = Y, *Ori = Mag + len, *W = Ori + len;
float *RBin = W + len, *CBin = RBin + len, *hist = CBin + len;
for( i = 0; i < d+2; i++ )
{
for( j = 0; j < d+2; j++ )
for( k = 0; k < n+2; k++ )
hist[(i*(d+2) + j)*(n+2) + k] = 0.;
}
for( i = -radius, k = 0; i <= radius; i++ )
for( j = -radius; j <= radius; j++ )
{
// Calculate sample's histogram array coords rotated relative to ori.
// Subtract 0.5 so samples that fall e.g. in the center of row 1 (i.e.
// r_rot = 1.5) have full weight placed in row 1 after interpolation.
float c_rot = j * cos_t - i * sin_t;
float r_rot = j * sin_t + i * cos_t;
float rbin = r_rot + d/2 - 0.5f;
float cbin = c_rot + d/2 - 0.5f;
int r = pt.y + i, c = pt.x + j;
if( rbin > -1 && rbin < d && cbin > -1 && cbin < d &&
r > 0 && r < rows - 1 && c > 0 && c < cols - 1 )
{
float dx = (float)(img.at<sift_wt>(r, c+1) - img.at<sift_wt>(r, c-1));
float dy = (float)(img.at<sift_wt>(r-1, c) - img.at<sift_wt>(r+1, c));
X[k] = dx; Y[k] = dy; RBin[k] = rbin; CBin[k] = cbin;
W[k] = (c_rot * c_rot + r_rot * r_rot)*exp_scale;
k++;
}
}
len = k;
cv::hal::fastAtan2(Y, X, Ori, len, true);
cv::hal::magnitude32f(X, Y, Mag, len);
cv::hal::exp32f(W, W, len);
k = 0;
#if CV_AVX2
if( USE_AVX2 )
{
int CV_DECL_ALIGNED(32) idx_buf[8];
float CV_DECL_ALIGNED(32) rco_buf[64];
const __m256 __ori = _mm256_set1_ps(ori);
const __m256 __bins_per_rad = _mm256_set1_ps(bins_per_rad);
const __m256i __n = _mm256_set1_epi32(n);
for( ; k <= len - 8; k+=8 )
{
__m256 __rbin = _mm256_loadu_ps(&RBin[k]);
__m256 __cbin = _mm256_loadu_ps(&CBin[k]);
__m256 __obin = _mm256_mul_ps(_mm256_sub_ps(_mm256_loadu_ps(&Ori[k]), __ori), __bins_per_rad);
__m256 __mag = _mm256_mul_ps(_mm256_loadu_ps(&Mag[k]), _mm256_loadu_ps(&W[k]));
__m256 __r0 = _mm256_floor_ps(__rbin);
__rbin = _mm256_sub_ps(__rbin, __r0);
__m256 __c0 = _mm256_floor_ps(__cbin);
__cbin = _mm256_sub_ps(__cbin, __c0);
__m256 __o0 = _mm256_floor_ps(__obin);
__obin = _mm256_sub_ps(__obin, __o0);
__m256i __o0i = _mm256_cvtps_epi32(__o0);
__o0i = _mm256_add_epi32(__o0i, _mm256_and_si256(__n, _mm256_cmpgt_epi32(_mm256_setzero_si256(), __o0i)));
__o0i = _mm256_sub_epi32(__o0i, _mm256_andnot_si256(_mm256_cmpgt_epi32(__n, __o0i), __n));
__m256 __v_r1 = _mm256_mul_ps(__mag, __rbin);
__m256 __v_r0 = _mm256_sub_ps(__mag, __v_r1);
__m256 __v_rc11 = _mm256_mul_ps(__v_r1, __cbin);
__m256 __v_rc10 = _mm256_sub_ps(__v_r1, __v_rc11);
__m256 __v_rc01 = _mm256_mul_ps(__v_r0, __cbin);
__m256 __v_rc00 = _mm256_sub_ps(__v_r0, __v_rc01);
__m256 __v_rco111 = _mm256_mul_ps(__v_rc11, __obin);
__m256 __v_rco110 = _mm256_sub_ps(__v_rc11, __v_rco111);
__m256 __v_rco101 = _mm256_mul_ps(__v_rc10, __obin);
__m256 __v_rco100 = _mm256_sub_ps(__v_rc10, __v_rco101);
__m256 __v_rco011 = _mm256_mul_ps(__v_rc01, __obin);
__m256 __v_rco010 = _mm256_sub_ps(__v_rc01, __v_rco011);
__m256 __v_rco001 = _mm256_mul_ps(__v_rc00, __obin);
__m256 __v_rco000 = _mm256_sub_ps(__v_rc00, __v_rco001);
__m256i __one = _mm256_set1_epi32(1);
__m256i __idx = _mm256_add_epi32(
_mm256_mullo_epi32(
_mm256_add_epi32(
_mm256_mullo_epi32(_mm256_add_epi32(_mm256_cvtps_epi32(__r0), __one), _mm256_set1_epi32(d + 2)),
_mm256_add_epi32(_mm256_cvtps_epi32(__c0), __one)),
_mm256_set1_epi32(n + 2)),
__o0i);
_mm256_store_si256((__m256i *)idx_buf, __idx);
_mm256_store_ps(&(rco_buf[0]), __v_rco000);
_mm256_store_ps(&(rco_buf[8]), __v_rco001);
_mm256_store_ps(&(rco_buf[16]), __v_rco010);
_mm256_store_ps(&(rco_buf[24]), __v_rco011);
_mm256_store_ps(&(rco_buf[32]), __v_rco100);
_mm256_store_ps(&(rco_buf[40]), __v_rco101);
_mm256_store_ps(&(rco_buf[48]), __v_rco110);
_mm256_store_ps(&(rco_buf[56]), __v_rco111);
#define HIST_SUM_HELPER(id) \
hist[idx_buf[(id)]] += rco_buf[(id)]; \
hist[idx_buf[(id)]+1] += rco_buf[8 + (id)]; \
hist[idx_buf[(id)]+(n+2)] += rco_buf[16 + (id)]; \
hist[idx_buf[(id)]+(n+3)] += rco_buf[24 + (id)]; \
hist[idx_buf[(id)]+(d+2)*(n+2)] += rco_buf[32 + (id)]; \
hist[idx_buf[(id)]+(d+2)*(n+2)+1] += rco_buf[40 + (id)]; \
hist[idx_buf[(id)]+(d+3)*(n+2)] += rco_buf[48 + (id)]; \
hist[idx_buf[(id)]+(d+3)*(n+2)+1] += rco_buf[56 + (id)];
HIST_SUM_HELPER(0);
HIST_SUM_HELPER(1);
HIST_SUM_HELPER(2);
HIST_SUM_HELPER(3);
HIST_SUM_HELPER(4);
HIST_SUM_HELPER(5);
HIST_SUM_HELPER(6);
HIST_SUM_HELPER(7);
#undef HIST_SUM_HELPER
}
}
#endif
for( ; k < len; k++ )
{
float rbin = RBin[k], cbin = CBin[k];
float obin = (Ori[k] - ori)*bins_per_rad;
float mag = Mag[k]*W[k];
int r0 = cvFloor( rbin );
int c0 = cvFloor( cbin );
int o0 = cvFloor( obin );
rbin -= r0;
cbin -= c0;
obin -= o0;
if( o0 < 0 )
o0 += n;
if( o0 >= n )
o0 -= n;
// histogram update using tri-linear interpolation
float v_r1 = mag*rbin, v_r0 = mag - v_r1;
float v_rc11 = v_r1*cbin, v_rc10 = v_r1 - v_rc11;
float v_rc01 = v_r0*cbin, v_rc00 = v_r0 - v_rc01;
float v_rco111 = v_rc11*obin, v_rco110 = v_rc11 - v_rco111;
float v_rco101 = v_rc10*obin, v_rco100 = v_rc10 - v_rco101;
float v_rco011 = v_rc01*obin, v_rco010 = v_rc01 - v_rco011;
float v_rco001 = v_rc00*obin, v_rco000 = v_rc00 - v_rco001;
int idx = ((r0+1)*(d+2) + c0+1)*(n+2) + o0;
hist[idx] += v_rco000;
hist[idx+1] += v_rco001;
hist[idx+(n+2)] += v_rco010;
hist[idx+(n+3)] += v_rco011;
hist[idx+(d+2)*(n+2)] += v_rco100;
hist[idx+(d+2)*(n+2)+1] += v_rco101;
hist[idx+(d+3)*(n+2)] += v_rco110;
hist[idx+(d+3)*(n+2)+1] += v_rco111;
}
// finalize histogram, since the orientation histograms are circular
for( i = 0; i < d; i++ )
for( j = 0; j < d; j++ )
{
int idx = ((i+1)*(d+2) + (j+1))*(n+2);
hist[idx] += hist[idx+n];
hist[idx+1] += hist[idx+n+1];
for( k = 0; k < n; k++ )
dst[(i*d + j)*n + k] = hist[idx+k];
}
// copy histogram to the descriptor,
// apply hysteresis thresholding
// and scale the result, so that it can be easily converted
// to byte array
float nrm2 = 0;
len = d*d*n;
k = 0;
#if CV_AVX2
if( USE_AVX2 )
{
float CV_DECL_ALIGNED(32) nrm2_buf[8];
__m256 __nrm2 = _mm256_setzero_ps();
__m256 __dst;
for( ; k <= len - 8; k += 8 )
{
__dst = _mm256_loadu_ps(&dst[k]);
#if CV_FMA3
__nrm2 = _mm256_fmadd_ps(__dst, __dst, __nrm2);
#else
__nrm2 = _mm256_add_ps(__nrm2, _mm256_mul_ps(__dst, __dst));
#endif
}
_mm256_store_ps(nrm2_buf, __nrm2);
nrm2 = nrm2_buf[0] + nrm2_buf[1] + nrm2_buf[2] + nrm2_buf[3] +
nrm2_buf[4] + nrm2_buf[5] + nrm2_buf[6] + nrm2_buf[7];
}
#endif
for( ; k < len; k++ )
nrm2 += dst[k]*dst[k];
float thr = std::sqrt(nrm2)*SIFT_DESCR_MAG_THR;
i = 0, nrm2 = 0;
#if 0 //CV_AVX2
// This code cannot be enabled because it sums nrm2 in a different order,
// thus producing slightly different results
if( USE_AVX2 )
{
float CV_DECL_ALIGNED(32) nrm2_buf[8];
__m256 __dst;
__m256 __nrm2 = _mm256_setzero_ps();
__m256 __thr = _mm256_set1_ps(thr);
for( ; i <= len - 8; i += 8 )
{
__dst = _mm256_loadu_ps(&dst[i]);
__dst = _mm256_min_ps(__dst, __thr);
_mm256_storeu_ps(&dst[i], __dst);
#if CV_FMA3
__nrm2 = _mm256_fmadd_ps(__dst, __dst, __nrm2);
#else
__nrm2 = _mm256_add_ps(__nrm2, _mm256_mul_ps(__dst, __dst));
#endif
}
_mm256_store_ps(nrm2_buf, __nrm2);
nrm2 = nrm2_buf[0] + nrm2_buf[1] + nrm2_buf[2] + nrm2_buf[3] +
nrm2_buf[4] + nrm2_buf[5] + nrm2_buf[6] + nrm2_buf[7];
}
#endif
for( ; i < len; i++ )
{
float val = std::min(dst[i], thr);
dst[i] = val;
nrm2 += val*val;
}
nrm2 = SIFT_INT_DESCR_FCTR/std::max(std::sqrt(nrm2), FLT_EPSILON);
#if 1
k = 0;
#if CV_AVX2
if( USE_AVX2 )
{
__m256 __dst;
__m256 __min = _mm256_setzero_ps();
__m256 __max = _mm256_set1_ps(255.0f); // max of uchar
__m256 __nrm2 = _mm256_set1_ps(nrm2);
for( k = 0; k <= len - 8; k+=8 )
{
__dst = _mm256_loadu_ps(&dst[k]);
__dst = _mm256_min_ps(_mm256_max_ps(_mm256_round_ps(_mm256_mul_ps(__dst, __nrm2), _MM_FROUND_TO_NEAREST_INT |_MM_FROUND_NO_EXC), __min), __max);
_mm256_storeu_ps(&dst[k], __dst);
}
}
#endif
for( ; k < len; k++ )
{
dst[k] = saturate_cast<uchar>(dst[k]*nrm2);
}
#else
float nrm1 = 0;
for( k = 0; k < len; k++ )
{
dst[k] *= nrm2;
nrm1 += dst[k];
}
nrm1 = 1.f/std::max(nrm1, FLT_EPSILON);
for( k = 0; k < len; k++ )
{
dst[k] = std::sqrt(dst[k] * nrm1);//saturate_cast<uchar>(std::sqrt(dst[k] * nrm1)*SIFT_INT_DESCR_FCTR);
}
#endif
}
class calcDescriptorsComputer : public ParallelLoopBody
{
public:
calcDescriptorsComputer(const std::vector<Mat>& _gpyr,
const std::vector<KeyPoint>& _keypoints,
Mat& _descriptors,
int _nOctaveLayers,
int _firstOctave)
: gpyr(_gpyr),
keypoints(_keypoints),
descriptors(_descriptors),
nOctaveLayers(_nOctaveLayers),
firstOctave(_firstOctave) { }
void operator()( const cv::Range& range ) const CV_OVERRIDE
{
const int begin = range.start;
const int end = range.end;
static const int d = SIFT_DESCR_WIDTH, n = SIFT_DESCR_HIST_BINS;