forked from sxyu/avatar
-
Notifications
You must be signed in to change notification settings - Fork 0
/
live-demo.cpp
534 lines (492 loc) · 22.8 KB
/
live-demo.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
#include <ctime>
#include <cstdlib>
#include <cstdio>
#include <chrono>
#include <string>
#include <vector>
#include <memory>
#include <algorithm>
#include <fstream>
#include <iostream>
#include <iomanip>
#include <Eigen/Dense>
#include <opencv2/core.hpp>
#include <opencv2/imgproc/imgproc.hpp>
#include <opencv2/highgui/highgui.hpp>
#include <boost/program_options.hpp>
#include <boost/filesystem.hpp>
#include <boost/program_options.hpp>
#include "Avatar.h"
#include "AvatarOptimizer.h"
#include "AvatarRenderer.h"
#include "BGSubtractor.h"
#include "Calibration.h"
#include "RTree.h"
#include "Util.h"
#ifdef OPENARK_AZURE_KINECT_ENABLED
#include "AzureKinectCamera.h"
#endif
#ifdef OPENARK_FREENECT2_ENABLED
#include "Freenect2Camera.h"
#endif
#include "opencv2/imgcodecs.hpp"
#define BEGIN_PROFILE // auto start = std::chrono::high_resolution_clock::now()
#define PROFILE( \
x) // do{printf("%s: %f ms\n", #x, std::chrono::duration<double,
// std::milli>(std::chrono::high_resolution_clock::now() -
// start).count()); start = std::chrono::high_resolution_clock::now();
// }while(false)
using namespace ark;
int main(int argc, char** argv) {
namespace po = boost::program_options;
std::string intrinPath, rtreePath, bgPath;
int nnStep, interval, frameICPIters, reinitICPIters, initialICPIters;
int initialPerPartCnz, reinitCnz, itersPerICP;
float betaPose, betaShape, nnDistThreshRel, neighbDistThreshRel,
distToPreWeight;
bool rtreeOnly, disableOcclusion;
cv::Size size;
bool forceK4a = false, forceFreenect2 = false;
po::options_description desc("Option arguments");
po::options_description descPositional("OpenARK Avatar Live Demo");
po::options_description descCombined("");
desc.add_options()("help", "produce help message")(
"rtree-only,R", po::bool_switch(&rtreeOnly),
"Show RTree part segmentation only and skip optimization")(
"no-occlusion", po::bool_switch(&disableOcclusion),
"Disable occlusion detection in avatar optimizer prior to NN matching")(
"betapose", po::value<float>(&betaPose)->default_value(0.05),
"Optimization loss function: pose prior term weight")(
"betashape", po::value<float>(&betaShape)->default_value(0.12),
"Optimization loss function: shape prior term weight")(
"nnstep", po::value<int>(&nnStep)->default_value(20),
"Optimization nearest-neighbor step: only matches neighbors every x "
"points; a heuristic to improve speed (currently, not used)")(
"data-interval,I", po::value<int>(&interval)->default_value(12),
"Only computes rtree weights and optimizes for pixels with x = y = 0 "
"mod interval")("frame-icp-iters,t",
po::value<int>(&frameICPIters)->default_value(3),
"ICP iterations per frame")(
"reinit-icp-iters,T", po::value<int>(&reinitICPIters)->default_value(5),
"ICP iterations when reinitializing (after tracking loss)")(
"initial-icp-iters,e",
po::value<int>(&initialICPIters)->default_value(7),
"ICP iterations when reinitializing (at beginning)")(
"inner-iters,p", po::value<int>(&itersPerICP)->default_value(10),
"Maximum inner iterations per ICP step")(
"intrin-path,i", po::value<std::string>(&intrinPath)->default_value(""),
"Path to camera intrinsics file (default: uses hardcoded K4A "
"intrinsics)")("bg-path,b",
po::value<std::string>(&bgPath)->default_value(""),
"Path to background image")(
"initial-per-part-thresh",
po::value<int>(&initialPerPartCnz)->default_value(80),
"Initial detected points per body part (/interval^2) to start tracking "
"avatar")(
"min-points,M", po::value<int>(&reinitCnz)->default_value(1000),
"Minimum number of detected body points to allow continued tracking; "
"if it falls below this number, then the tracker reinitializes")(
"nn-dist,N", po::value<float>(&nnDistThreshRel)->default_value(0.002),
"Min distance (scales with image size) between background pixel and "
"current pixel")(
"neighb-dist,n",
po::value<float>(&neighbDistThreshRel)->default_value(0.001),
"Min distance (scales with image size) between neighboring pixels in "
"same component, for BG subtraction")(
"dist-to-pre-weight",
po::value<float>(&distToPreWeight)->default_value(0.001),
"Weight of squared to distance to previous center of mass when "
"choosing best connected component for each body part (RTree "
"postprocessing)")("width",
po::value<int>(&size.width)->default_value(1280),
"Width of generated images")(
"height", po::value<int>(&size.height)->default_value(720),
"Height of generated imaes")
#ifdef OPENARK_AZURE_KINECT_ENABLED
("k4a", po::bool_switch(&forceK4a),
"if set, forces Kinect Azure (k4a) depth camera")
#endif
#ifdef OPENARK_FREENECT2_ENABLED
("freenect2", po::bool_switch(&forceFreenect2),
"if set, forces Freenect2 depth camera")
#endif
;
descPositional.add_options()("rtree",
po::value<std::string>(&rtreePath)->required(),
"RTree model path");
descCombined.add(descPositional);
descCombined.add(desc);
po::variables_map vm;
po::positional_options_description posopt;
posopt.add("rtree", 1);
try {
po::store(po::command_line_parser(argc, argv)
.options(descCombined)
.positional(posopt)
.run(),
vm);
} catch (std::exception& e) {
std::cerr << "Error: " << e.what() << "\n";
std::cerr << descPositional << "\n" << desc << "\n";
return 1;
}
if (vm.count("help")) {
std::cout << descPositional << "\n" << desc << "\n";
return 0;
}
try {
po::notify(vm);
} catch (std::exception& e) {
std::cerr << "Error: " << e.what() << "\n";
std::cerr << descPositional << "\n" << desc << "\n";
return 1;
}
if ((int)forceK4a + (int)forceFreenect2 > 1) {
std::cerr << "Only one camera preference may be provided";
return 1;
}
printf(
"CONTROLS:\nQ or ESC to quit\n"
"b to set background (also sets on first unpause, if -b not "
"specified)\n"
"0-3 to show empty, RGB, depth, custom (ext_background.jpg) "
"background behind avatar\n"
"h to hide/show human bounding box from background subtraction\n"
"t to toggle random tree visualization/avatar tracking visualization\n"
"SPACE to start/pause\n\n");
// Seed the rng
srand(time(NULL));
CameraIntrin intrin;
if (!intrinPath.empty())
intrin.readFile(intrinPath);
else {
intrin.clear();
intrin.fx = 606.438;
intrin.fy = 606.351;
intrin.cx = 637.294;
intrin.cy = 366.992;
}
_ARK_ASSERT((bool)std::ifstream(rtreePath));
ark::RTree rtree(0);
rtree.loadFile(rtreePath);
ark::AvatarModel avaModel;
ark::AvatarModel avaModelCheap("", false);
ark::Avatar avaFull(avaModel);
ark::Avatar ava(avaModelCheap);
ark::AvatarOptimizer avaOpt(ava, intrin, size, rtree.numParts,
rtree.partMap);
avaOpt.betaPose = betaPose;
avaOpt.betaShape = betaShape;
avaOpt.nnStep = nnStep;
avaOpt.enableOcclusion = !disableOcclusion;
avaOpt.maxItersPerICP = itersPerICP;
ark::BGSubtractor bgsub{cv::Mat()};
bgsub.numThreads = std::thread::hardware_concurrency();
bgsub.nnDistThreshRel = nnDistThreshRel;
bgsub.neighbThreshRel = neighbDistThreshRel;
if (bgPath.size()) {
util::readXYZ(bgPath, bgsub.background, intrin);
}
// Background image #3, any image at ./ext_background.jpg (optional)
cv::Mat extBackgroundImage = cv::imread("ext_background.jpg");
// Previous centers of mass: required by RTree postprocessor
Eigen::Matrix<double, 2, Eigen::Dynamic> comPre;
// Initialize the camera
DepthCamera::Ptr camera;
if (forceK4a) {
#ifdef OPENARK_AZURE_KINECT_ENABLED
camera = std::make_shared<AzureKinectCamera>();
#endif
} else if (forceFreenect2) {
#ifdef OPENARK_FREENECT2_ENABLED
camera = std::make_shared<Freenect2Camera>();
#endif
} else {
camera = std::make_shared<OPENARK_PREFERRED_CAMERA>();
}
auto capture_start_time = std::chrono::high_resolution_clock::now();
// Turn on the camera
camera->beginCapture();
if (!camera->isCapturing()) {
std::cerr << "Failed to open camera, quitting...\n";
return 1;
}
// Read in camera input and save it to the buffer
std::vector<uint64_t> timestamps;
// Pausing feature: if true, demo is paused
bool pause = true;
// Background type: 0 = none 1 = RGB 2 = depth 3 = ./ext_background.jpg
int backgroundType = 1;
// If true, shows bounding box of BG subtraction area
bool showBoundingBox = rtreeOnly;
// When this flag is true, tracking will reinit the next frame
bool reinit = true;
// This indicates if we are reinitializing for the first time
bool firstTime = true;
std::cerr << "Note: paused, press space to begin recording.\n"
"The background (for BG subtraction) will be captured "
"automatically each time you unpause.\nPlease stay out of the "
"grayish area (where depth is unavailable) if possible.\n";
int currFrame = 0; // current frame number (since launch/last pause)
while (true) {
++currFrame;
// get latest image from the camera
cv::Mat xyzMap = camera->getXYZMap();
cv::Mat rgbMap = camera->getRGBMap();
if (!xyzMap.empty() && !rgbMap.empty()) {
if (pause) {
const cv::Scalar RECT_COLOR = cv::Scalar(0, 160, 255);
const std::string NO_SIGNAL_STR = "PAUSED";
const cv::Point STR_POS(rgbMap.cols / 2 - 50,
rgbMap.rows / 2 + 7);
const int RECT_WID = 120, RECT_HI = 40;
cv::Rect rect(rgbMap.cols / 2 - RECT_WID / 2,
rgbMap.rows / 2 - RECT_HI / 2, RECT_WID, RECT_HI);
// show 'paused' and do not record
cv::rectangle(rgbMap, rect, RECT_COLOR, -1);
cv::putText(rgbMap, NO_SIGNAL_STR, STR_POS, 0, 0.8,
cv::Scalar(255, 255, 255), 1, cv::LINE_AA);
std::this_thread::sleep_for(std::chrono::milliseconds(100));
// cv::rectangle(xyzMap, rect, RECT_COLOR / 255.0, -1);
// cv::putText(xyzMap, NO_SIGNAL_STR, STR_POS, 0, 0.8,
// cv::Scalar(1.0f, 1.0f, 1.0f), 1, cv::LINE_AA);
} else {
// Capture image if timestamp changed
auto ts = camera->getTimestamp();
if (timestamps.size() && ts - timestamps.back() < 100000) {
continue;
}
}
cv::Mat depth;
cv::extractChannel(xyzMap, depth, 2);
// Replace RGB stream with user-specified background
if (backgroundType == 0) {
rgbMap = cv::Mat::zeros(rgbMap.size(), CV_8UC3);
} else if (backgroundType == 2) {
xyzMap.convertTo(rgbMap, CV_8UC3, 255.);
} else if (backgroundType == 3) {
if (extBackgroundImage.empty()) {
extBackgroundImage = cv::Mat::zeros(rgbMap.size(), CV_8UC3);
} else if (extBackgroundImage.size() != rgbMap.size()) {
cv::resize(extBackgroundImage, extBackgroundImage,
rgbMap.size());
}
extBackgroundImage.copyTo(rgbMap);
}
cv::Mat visual, rgbMapFloat;
if (!pause) {
if (!bgsub.background.empty()) {
cv::Mat sub = bgsub.run(xyzMap);
size_t subCnz = 0;
for (int r = bgsub.topLeft.y; r <= bgsub.botRight.y; ++r) {
// auto* outptr = rgbMap.ptr<cv::Vec3b>(r);
const auto* inptr = sub.ptr<uint8_t>(r);
auto* dptr = depth.ptr<float>(r);
for (int c = bgsub.topLeft.x; c <= bgsub.botRight.x;
++c) {
int colorIdx = inptr[c];
if (colorIdx >= 254) {
dptr[c] = 0;
} else {
++subCnz;
}
}
}
if (subCnz < reinitCnz / (interval * interval)) {
if (reinit == false) {
std::cout << "Note: detected empty scene, "
"decreasing frame rate\n";
reinit = true;
}
std::this_thread::sleep_for(
std::chrono::milliseconds(100));
} else {
cv::Mat result = rtree.predictBest(
depth, std::thread::hardware_concurrency(), 2,
bgsub.topLeft, bgsub.botRight);
if (rtreeOnly) {
for (int r = bgsub.topLeft.y; r <= bgsub.botRight.y;
++r) {
auto* inPtr = result.ptr<uint8_t>(r);
auto* visualPtr = rgbMap.ptr<cv::Vec3b>(r);
for (int c = bgsub.topLeft.x;
c <= bgsub.botRight.x; ++c) {
if (inPtr[c] == 255) continue;
visualPtr[c] =
ark::util::paletteColor(inPtr[c], true);
}
}
} else {
rtree.postProcess(
result, comPre, 2,
std::thread::hardware_concurrency(),
bgsub.topLeft, bgsub.botRight, distToPreWeight);
Eigen::Matrix<size_t, Eigen::Dynamic, 1> partCnz(
rtree.numParts);
partCnz.setZero();
for (int r = bgsub.topLeft.y; r <= bgsub.botRight.y;
r += interval) {
auto* partptr = result.ptr<uint8_t>(r);
for (int c = bgsub.topLeft.x;
c <= bgsub.botRight.x; c += interval) {
if (partptr[c] == 255) continue;
++partCnz[partptr[c]];
}
}
size_t cnz = partCnz.sum();
if ((firstTime &&
partCnz.minCoeff() <
std::max(1, initialPerPartCnz /
(interval * interval))) ||
cnz < reinitCnz / (interval * interval)) {
reinit = true;
} else {
ark::CloudType dataCloud(3, cnz);
Eigen::VectorXi dataPartLabels(cnz);
size_t i = 0;
for (int r = bgsub.topLeft.y;
r <= bgsub.botRight.y; r += interval) {
auto* ptr = xyzMap.ptr<cv::Vec3f>(r);
auto* partptr = result.ptr<uint8_t>(r);
for (int c = bgsub.topLeft.x;
c <= bgsub.botRight.x; c += interval) {
if (partptr[c] == 255) continue;
dataCloud(0, i) = ptr[c][0];
dataCloud(1, i) = -ptr[c][1];
dataCloud(2, i) = ptr[c][2];
dataPartLabels(i) = partptr[c];
++i;
}
}
int icpIters = frameICPIters;
if (reinit) {
Eigen::Vector3d cloudCen =
dataCloud.rowwise().mean();
ava.p = cloudCen;
ava.w.setZero();
for (int i = 1; i < ava.model.numJoints();
++i) {
ava.r[i].setIdentity();
}
ava.r[0] =
Eigen::AngleAxisd(
M_PI, Eigen::Vector3d(0, 1, 0))
.toRotationMatrix();
reinit = false;
ava.update();
icpIters = firstTime ? initialICPIters
: reinitICPIters;
std::cerr
<< "Note: reinitializing tracking\n";
if (firstTime) firstTime = false;
}
BEGIN_PROFILE;
avaOpt.optimize(
dataCloud, dataPartLabels, icpIters,
std::thread::hardware_concurrency());
PROFILE(Optimize(Total));
ark::AvatarRenderer rend(ava, intrin);
avaFull.r = ava.r;
avaFull.w = ava.w;
avaFull.p = ava.p;
avaFull.update();
cv::Mat modelMap =
rend.renderLambert(depth.size());
for (int r = 0; r < rgbMap.rows; ++r) {
auto* outptr = rgbMap.ptr<cv::Vec3b>(r);
const auto* renderptr =
modelMap.ptr<uint8_t>(r);
for (int c = 0; c < rgbMap.cols; ++c) {
if (renderptr[c] > 0) {
if (backgroundType == 1 ||
backgroundType == 2) {
outptr[c] =
cv::Vec3b(renderptr[c],
renderptr[c],
renderptr[c]) /
5 * 3 +
outptr[c] / 5 * 2;
} else {
outptr[c] = cv::Vec3b(
renderptr[c], renderptr[c],
renderptr[c]);
}
}
}
}
}
}
}
}
}
rgbMap.convertTo(rgbMapFloat, CV_32FC3, 1. / 255.);
// cv::hconcat(xyzMap, rgbMapFloat, visual);
visual = rgbMapFloat;
if (backgroundType == 1) {
for (int r = 0; r < visual.rows; ++r) {
auto* outptr = visual.ptr<cv::Vec3f>(r);
auto* xyzptr = xyzMap.ptr<cv::Vec3f>(r);
for (int c = 0; c < visual.cols; ++c) {
if (xyzptr[c][2] == 0.0f) {
// Since depth camera's FoV is actually
// smaller than RGB FoV visible,
// try to make user aware of
// depth boundaries by darkening areas with
// no depth info
outptr[c] *= 0.5;
}
}
}
}
const int MAX_COLS = 1300;
if (showBoundingBox) {
cv::rectangle(visual, bgsub.topLeft, bgsub.botRight,
cv::Scalar(0, 0, 255));
}
if (visual.cols > MAX_COLS) {
cv::resize(
visual, visual,
cv::Size(MAX_COLS, MAX_COLS * visual.rows / visual.cols));
}
cv::imshow(camera->getModelName() + " Results", visual);
}
int c = cv::waitKey(1);
// make case insensitive (convert to upper)
if (c >= 'a' && c <= 'z') c &= 0xdf;
if (c == 'Q' || c == 27) {
// 27 is ESC
break;
}
if (c >= '0' && c <= '3') {
backgroundType = c - '0';
std::cout << "Background: " << backgroundType << "\n";
}
switch (c) {
case 'B':
bgsub.background = xyzMap;
std::cout << "Note: background updated.\n";
break;
case 'H':
showBoundingBox = !showBoundingBox;
std::cout << "Bounding box visible: " << std::boolalpha
<< showBoundingBox << "." << std::endl;
break;
case 'T':
rtreeOnly = !rtreeOnly;
std::cout << "Random tree visulization mode: " << std::boolalpha
<< rtreeOnly << "." << std::endl;
break;
case ' ':
if (bgsub.background.empty()) {
bgsub.background = xyzMap;
std::cout << "Note: unpaused, background updated.\n";
}
pause = !pause;
if (pause) reinit = true;
break;
}
}
camera->endCapture();
cv::destroyAllWindows();
return 0;
}