| 1 | // This file is part of OpenCV project. |
| 2 | // It is subject to the license terms in the LICENSE file found in the top-level directory |
| 3 | // of this distribution and at http://opencv.org/license.html. |
| 4 | |
| 5 | #include "../precomp.hpp" |
| 6 | |
| 7 | #ifdef HAVE_OPENCV_DNN |
| 8 | #include "opencv2/dnn.hpp" |
| 9 | #endif |
| 10 | |
| 11 | namespace cv { |
| 12 | |
| 13 | TrackerDaSiamRPN::TrackerDaSiamRPN() |
| 14 | { |
| 15 | // nothing |
| 16 | } |
| 17 | |
| 18 | TrackerDaSiamRPN::~TrackerDaSiamRPN() |
| 19 | { |
| 20 | // nothing |
| 21 | } |
| 22 | |
| 23 | TrackerDaSiamRPN::Params::Params() |
| 24 | { |
| 25 | model = "dasiamrpn_model.onnx" ; |
| 26 | kernel_cls1 = "dasiamrpn_kernel_cls1.onnx" ; |
| 27 | kernel_r1 = "dasiamrpn_kernel_r1.onnx" ; |
| 28 | #ifdef HAVE_OPENCV_DNN |
| 29 | backend = dnn::DNN_BACKEND_DEFAULT; |
| 30 | target = dnn::DNN_TARGET_CPU; |
| 31 | #else |
| 32 | backend = -1; // invalid value |
| 33 | target = -1; // invalid value |
| 34 | #endif |
| 35 | } |
| 36 | |
| 37 | #ifdef HAVE_OPENCV_DNN |
| 38 | |
| 39 | template <typename T> static |
| 40 | T sizeCal(const T& w, const T& h) |
| 41 | { |
| 42 | T pad = (w + h) * T(0.5); |
| 43 | T sz2 = (w + pad) * (h + pad); |
| 44 | return sqrt(sz2); |
| 45 | } |
| 46 | |
| 47 | template <> |
| 48 | Mat sizeCal(const Mat& w, const Mat& h) |
| 49 | { |
| 50 | Mat pad = (w + h) * 0.5; |
| 51 | Mat sz2 = (w + pad).mul(e: (h + pad)); |
| 52 | |
| 53 | cv::sqrt(src: sz2, dst: sz2); |
| 54 | return sz2; |
| 55 | } |
| 56 | |
| 57 | class TrackerDaSiamRPNImpl : public TrackerDaSiamRPN |
| 58 | { |
| 59 | public: |
| 60 | TrackerDaSiamRPNImpl(const TrackerDaSiamRPN::Params& params) |
| 61 | { |
| 62 | siamRPN = dnn::readNet(model: params.model); |
| 63 | siamKernelCL1 = dnn::readNet(model: params.kernel_cls1); |
| 64 | siamKernelR1 = dnn::readNet(model: params.kernel_r1); |
| 65 | |
| 66 | CV_Assert(!siamRPN.empty()); |
| 67 | CV_Assert(!siamKernelCL1.empty()); |
| 68 | CV_Assert(!siamKernelR1.empty()); |
| 69 | |
| 70 | siamRPN.setPreferableBackend(params.backend); |
| 71 | siamRPN.setPreferableTarget(params.target); |
| 72 | siamKernelR1.setPreferableBackend(params.backend); |
| 73 | siamKernelR1.setPreferableTarget(params.target); |
| 74 | siamKernelCL1.setPreferableBackend(params.backend); |
| 75 | siamKernelCL1.setPreferableTarget(params.target); |
| 76 | } |
| 77 | |
| 78 | TrackerDaSiamRPNImpl(const dnn::Net& siam_rpn, const dnn::Net& kernel_cls1, const dnn::Net& kernel_r1) |
| 79 | { |
| 80 | CV_Assert(!siam_rpn.empty()); |
| 81 | CV_Assert(!kernel_cls1.empty()); |
| 82 | CV_Assert(!kernel_r1.empty()); |
| 83 | |
| 84 | siamRPN = siam_rpn; |
| 85 | siamKernelCL1 = kernel_cls1; |
| 86 | siamKernelR1 = kernel_r1; |
| 87 | } |
| 88 | |
| 89 | void init(InputArray image, const Rect& boundingBox) CV_OVERRIDE; |
| 90 | bool update(InputArray image, Rect& boundingBox) CV_OVERRIDE; |
| 91 | float getTrackingScore() CV_OVERRIDE; |
| 92 | |
| 93 | protected: |
| 94 | dnn::Net siamRPN, siamKernelR1, siamKernelCL1; |
| 95 | Rect boundingBox_; |
| 96 | Mat image_; |
| 97 | struct trackerConfig |
| 98 | { |
| 99 | float windowInfluence = 0.43f; |
| 100 | float lr = 0.4f; |
| 101 | int scale = 8; |
| 102 | bool swapRB = false; |
| 103 | int totalStride = 8; |
| 104 | float penaltyK = 0.055f; |
| 105 | int exemplarSize = 127; |
| 106 | int instanceSize = 271; |
| 107 | float contextAmount = 0.5f; |
| 108 | std::vector<float> ratios = { 0.33f, 0.5f, 1.0f, 2.0f, 3.0f }; |
| 109 | int anchorNum = int(ratios.size()); |
| 110 | Mat anchors; |
| 111 | Mat windows; |
| 112 | Scalar avgChans; |
| 113 | Size imgSize = { 0, 0 }; |
| 114 | Rect2f targetBox = { 0, 0, 0, 0 }; |
| 115 | int scoreSize = (instanceSize - exemplarSize) / totalStride + 1; |
| 116 | float tracking_score; |
| 117 | |
| 118 | void update_scoreSize() |
| 119 | { |
| 120 | scoreSize = int((instanceSize - exemplarSize) / totalStride + 1); |
| 121 | } |
| 122 | }; |
| 123 | trackerConfig trackState; |
| 124 | |
| 125 | void softmax(const Mat& src, Mat& dst); |
| 126 | void elementMax(Mat& src); |
| 127 | Mat generateHanningWindow(); |
| 128 | Mat generateAnchors(); |
| 129 | Mat getSubwindow(Mat& img, const Rect2f& targetBox, float originalSize, Scalar avgChans); |
| 130 | void trackerInit(Mat img); |
| 131 | void trackerEval(Mat img); |
| 132 | }; |
| 133 | |
| 134 | void TrackerDaSiamRPNImpl::init(InputArray image, const Rect& boundingBox) |
| 135 | { |
| 136 | image_ = image.getMat().clone(); |
| 137 | |
| 138 | trackState.update_scoreSize(); |
| 139 | trackState.targetBox = Rect2f( |
| 140 | float(boundingBox.x) + float(boundingBox.width) * 0.5f, // FIXIT don't use center in Rect structures, it is confusing |
| 141 | float(boundingBox.y) + float(boundingBox.height) * 0.5f, |
| 142 | float(boundingBox.width), |
| 143 | float(boundingBox.height) |
| 144 | ); |
| 145 | trackerInit(img: image_); |
| 146 | } |
| 147 | |
| 148 | void TrackerDaSiamRPNImpl::trackerInit(Mat img) |
| 149 | { |
| 150 | Rect2f targetBox = trackState.targetBox; |
| 151 | Mat anchors = generateAnchors(); |
| 152 | trackState.anchors = anchors; |
| 153 | |
| 154 | Mat windows = generateHanningWindow(); |
| 155 | |
| 156 | trackState.windows = windows; |
| 157 | trackState.imgSize = img.size(); |
| 158 | |
| 159 | trackState.avgChans = mean(src: img); |
| 160 | float wc = targetBox.width + trackState.contextAmount * (targetBox.width + targetBox.height); |
| 161 | float hc = targetBox.height + trackState.contextAmount * (targetBox.width + targetBox.height); |
| 162 | float sz = (float)cvRound(value: sqrt(x: wc * hc)); |
| 163 | |
| 164 | Mat zCrop = getSubwindow(img, targetBox, originalSize: sz, avgChans: trackState.avgChans); |
| 165 | Mat blob; |
| 166 | |
| 167 | dnn::blobFromImage(image: zCrop, blob, scalefactor: 1.0, size: Size(trackState.exemplarSize, trackState.exemplarSize), mean: Scalar(), swapRB: trackState.swapRB, crop: false, CV_32F); |
| 168 | siamRPN.setInput(blob); |
| 169 | Mat out1; |
| 170 | siamRPN.forward(outputBlobs: out1, outputName: "onnx_node_output_0!63" ); |
| 171 | |
| 172 | siamKernelCL1.setInput(blob: out1); |
| 173 | siamKernelR1.setInput(blob: out1); |
| 174 | |
| 175 | Mat cls1 = siamKernelCL1.forward(); |
| 176 | Mat r1 = siamKernelR1.forward(); |
| 177 | std::vector<int> r1_shape = { 20, 256, 4, 4 }, cls1_shape = { 10, 256, 4, 4 }; |
| 178 | |
| 179 | siamRPN.setParam(layer: siamRPN.getLayerId(layer: "onnx_node_output_0!65" ), numParam: 0, blob: r1.reshape(cn: 0, newshape: r1_shape)); |
| 180 | siamRPN.setParam(layer: siamRPN.getLayerId(layer: "onnx_node_output_0!68" ), numParam: 0, blob: cls1.reshape(cn: 0, newshape: cls1_shape)); |
| 181 | } |
| 182 | |
| 183 | bool TrackerDaSiamRPNImpl::update(InputArray image, Rect& boundingBox) |
| 184 | { |
| 185 | image_ = image.getMat().clone(); |
| 186 | trackerEval(img: image_); |
| 187 | boundingBox = { |
| 188 | int(trackState.targetBox.x - int(trackState.targetBox.width / 2)), |
| 189 | int(trackState.targetBox.y - int(trackState.targetBox.height / 2)), |
| 190 | int(trackState.targetBox.width), |
| 191 | int(trackState.targetBox.height) |
| 192 | }; |
| 193 | return true; |
| 194 | } |
| 195 | |
| 196 | void TrackerDaSiamRPNImpl::trackerEval(Mat img) |
| 197 | { |
| 198 | Rect2f targetBox = trackState.targetBox; |
| 199 | |
| 200 | float wc = targetBox.height + trackState.contextAmount * (targetBox.width + targetBox.height); |
| 201 | float hc = targetBox.width + trackState.contextAmount * (targetBox.width + targetBox.height); |
| 202 | |
| 203 | float sz = sqrt(x: wc * hc); |
| 204 | float scaleZ = trackState.exemplarSize / sz; |
| 205 | |
| 206 | float searchSize = float((trackState.instanceSize - trackState.exemplarSize) / 2); |
| 207 | float pad = searchSize / scaleZ; |
| 208 | float sx = sz + 2 * pad; |
| 209 | |
| 210 | Mat xCrop = getSubwindow(img, targetBox, originalSize: (float)cvRound(value: sx), avgChans: trackState.avgChans); |
| 211 | |
| 212 | Mat blob; |
| 213 | std::vector<Mat> outs; |
| 214 | std::vector<String> outNames; |
| 215 | Mat delta, score; |
| 216 | Mat sc, rc, penalty, pscore; |
| 217 | |
| 218 | dnn::blobFromImage(image: xCrop, blob, scalefactor: 1.0, size: Size(trackState.instanceSize, trackState.instanceSize), mean: Scalar(), swapRB: trackState.swapRB, crop: false, CV_32F); |
| 219 | |
| 220 | siamRPN.setInput(blob); |
| 221 | |
| 222 | outNames = siamRPN.getUnconnectedOutLayersNames(); |
| 223 | siamRPN.forward(outputBlobs: outs, outBlobNames: outNames); |
| 224 | |
| 225 | delta = outs[0]; |
| 226 | score = outs[1]; |
| 227 | |
| 228 | score = score.reshape(cn: 0, newshape: { 2, trackState.anchorNum, trackState.scoreSize, trackState.scoreSize }); |
| 229 | delta = delta.reshape(cn: 0, newshape: { 4, trackState.anchorNum, trackState.scoreSize, trackState.scoreSize }); |
| 230 | |
| 231 | softmax(src: score, dst&: score); |
| 232 | |
| 233 | targetBox.width *= scaleZ; |
| 234 | targetBox.height *= scaleZ; |
| 235 | |
| 236 | score = score.row(y: 1); |
| 237 | score = score.reshape(cn: 0, newshape: { 5, 19, 19 }); |
| 238 | |
| 239 | // Post processing |
| 240 | delta.row(y: 0) = delta.row(y: 0).mul(m: trackState.anchors.row(y: 2)) + trackState.anchors.row(y: 0); |
| 241 | delta.row(y: 1) = delta.row(y: 1).mul(m: trackState.anchors.row(y: 3)) + trackState.anchors.row(y: 1); |
| 242 | exp(src: delta.row(y: 2), dst: delta.row(y: 2)); |
| 243 | delta.row(y: 2) = delta.row(y: 2).mul(m: trackState.anchors.row(y: 2)); |
| 244 | exp(src: delta.row(y: 3), dst: delta.row(y: 3)); |
| 245 | delta.row(y: 3) = delta.row(y: 3).mul(m: trackState.anchors.row(y: 3)); |
| 246 | |
| 247 | sc = sizeCal(w: delta.row(y: 2), h: delta.row(y: 3)) / sizeCal(w: targetBox.width, h: targetBox.height); |
| 248 | elementMax(src&: sc); |
| 249 | |
| 250 | rc = delta.row(y: 2).mul(m: 1 / delta.row(y: 3)); |
| 251 | rc = (targetBox.width / targetBox.height) / rc; |
| 252 | elementMax(src&: rc); |
| 253 | |
| 254 | // Calculating the penalty |
| 255 | exp(src: ((rc.mul(m: sc) - 1.) * trackState.penaltyK * (-1.0)), dst: penalty); |
| 256 | penalty = penalty.reshape(cn: 0, newshape: { trackState.anchorNum, trackState.scoreSize, trackState.scoreSize }); |
| 257 | |
| 258 | pscore = penalty.mul(m: score); |
| 259 | pscore = pscore * (1.0 - trackState.windowInfluence) + trackState.windows * trackState.windowInfluence; |
| 260 | |
| 261 | int bestID[2] = { 0, 0 }; |
| 262 | // Find the index of best score. |
| 263 | minMaxIdx(src: pscore.reshape(cn: 0, newshape: { trackState.anchorNum * trackState.scoreSize * trackState.scoreSize, 1 }), minVal: 0, maxVal: 0, minIdx: 0, maxIdx: bestID); |
| 264 | delta = delta.reshape(cn: 0, newshape: { 4, trackState.anchorNum * trackState.scoreSize * trackState.scoreSize }); |
| 265 | penalty = penalty.reshape(cn: 0, newshape: { trackState.anchorNum * trackState.scoreSize * trackState.scoreSize, 1 }); |
| 266 | score = score.reshape(cn: 0, newshape: { trackState.anchorNum * trackState.scoreSize * trackState.scoreSize, 1 }); |
| 267 | |
| 268 | int index[2] = { 0, bestID[0] }; |
| 269 | Rect2f resBox = { 0, 0, 0, 0 }; |
| 270 | |
| 271 | resBox.x = delta.at<float>(idx: index) / scaleZ; |
| 272 | index[0] = 1; |
| 273 | resBox.y = delta.at<float>(idx: index) / scaleZ; |
| 274 | index[0] = 2; |
| 275 | resBox.width = delta.at<float>(idx: index) / scaleZ; |
| 276 | index[0] = 3; |
| 277 | resBox.height = delta.at<float>(idx: index) / scaleZ; |
| 278 | |
| 279 | float lr = penalty.at<float>(idx: bestID) * score.at<float>(idx: bestID) * trackState.lr; |
| 280 | |
| 281 | resBox.x = resBox.x + targetBox.x; |
| 282 | resBox.y = resBox.y + targetBox.y; |
| 283 | targetBox.width /= scaleZ; |
| 284 | targetBox.height /= scaleZ; |
| 285 | |
| 286 | resBox.width = targetBox.width * (1 - lr) + resBox.width * lr; |
| 287 | resBox.height = targetBox.height * (1 - lr) + resBox.height * lr; |
| 288 | |
| 289 | resBox.x = float(fmax(x: 0., y: fmin(x: float(trackState.imgSize.width), y: resBox.x))); |
| 290 | resBox.y = float(fmax(x: 0., y: fmin(x: float(trackState.imgSize.height), y: resBox.y))); |
| 291 | resBox.width = float(fmax(x: 10., y: fmin(x: float(trackState.imgSize.width), y: resBox.width))); |
| 292 | resBox.height = float(fmax(x: 10., y: fmin(x: float(trackState.imgSize.height), y: resBox.height))); |
| 293 | |
| 294 | trackState.targetBox = resBox; |
| 295 | trackState.tracking_score = score.at<float>(idx: bestID); |
| 296 | } |
| 297 | |
| 298 | float TrackerDaSiamRPNImpl::getTrackingScore() |
| 299 | { |
| 300 | return trackState.tracking_score; |
| 301 | } |
| 302 | |
| 303 | void TrackerDaSiamRPNImpl::softmax(const Mat& src, Mat& dst) |
| 304 | { |
| 305 | Mat maxVal; |
| 306 | cv::max(src1: src.row(y: 1), src2: src.row(y: 0), dst&: maxVal); |
| 307 | |
| 308 | src.row(y: 1) -= maxVal; |
| 309 | src.row(y: 0) -= maxVal; |
| 310 | |
| 311 | exp(src, dst); |
| 312 | |
| 313 | Mat sumVal = dst.row(y: 0) + dst.row(y: 1); |
| 314 | dst.row(y: 0) = dst.row(y: 0) / sumVal; |
| 315 | dst.row(y: 1) = dst.row(y: 1) / sumVal; |
| 316 | } |
| 317 | |
| 318 | void TrackerDaSiamRPNImpl::elementMax(Mat& src) |
| 319 | { |
| 320 | int* p = src.size.p; |
| 321 | int index[4] = { 0, 0, 0, 0 }; |
| 322 | for (int n = 0; n < *p; n++) |
| 323 | { |
| 324 | for (int k = 0; k < *(p + 1); k++) |
| 325 | { |
| 326 | for (int i = 0; i < *(p + 2); i++) |
| 327 | { |
| 328 | for (int j = 0; j < *(p + 3); j++) |
| 329 | { |
| 330 | index[0] = n, index[1] = k, index[2] = i, index[3] = j; |
| 331 | float& v = src.at<float>(idx: index); |
| 332 | v = fmax(x: v, y: 1.0f / v); |
| 333 | } |
| 334 | } |
| 335 | } |
| 336 | } |
| 337 | } |
| 338 | |
| 339 | Mat TrackerDaSiamRPNImpl::generateHanningWindow() |
| 340 | { |
| 341 | Mat baseWindows, HanningWindows; |
| 342 | |
| 343 | createHanningWindow(dst: baseWindows, winSize: Size(trackState.scoreSize, trackState.scoreSize), CV_32F); |
| 344 | baseWindows = baseWindows.reshape(cn: 0, newshape: { 1, trackState.scoreSize, trackState.scoreSize }); |
| 345 | HanningWindows = baseWindows.clone(); |
| 346 | for (int i = 1; i < trackState.anchorNum; i++) |
| 347 | { |
| 348 | HanningWindows.push_back(m: baseWindows); |
| 349 | } |
| 350 | |
| 351 | return HanningWindows; |
| 352 | } |
| 353 | |
| 354 | Mat TrackerDaSiamRPNImpl::generateAnchors() |
| 355 | { |
| 356 | int totalStride = trackState.totalStride, scales = trackState.scale, scoreSize = trackState.scoreSize; |
| 357 | std::vector<float> ratios = trackState.ratios; |
| 358 | std::vector<Rect2f> baseAnchors; |
| 359 | int anchorNum = int(ratios.size()); |
| 360 | int size = totalStride * totalStride; |
| 361 | |
| 362 | float ori = -(float(scoreSize / 2)) * float(totalStride); |
| 363 | |
| 364 | for (auto i = 0; i < anchorNum; i++) |
| 365 | { |
| 366 | int ws = int(sqrt(x: size / ratios[i])); |
| 367 | int hs = int(ws * ratios[i]); |
| 368 | |
| 369 | float wws = float(ws) * scales; |
| 370 | float hhs = float(hs) * scales; |
| 371 | Rect2f anchor = { 0, 0, wws, hhs }; |
| 372 | baseAnchors.push_back(x: anchor); |
| 373 | } |
| 374 | |
| 375 | int anchorIndex[4] = { 0, 0, 0, 0 }; |
| 376 | const int sizes[4] = { 4, (int)ratios.size(), scoreSize, scoreSize }; |
| 377 | Mat anchors(4, sizes, CV_32F); |
| 378 | |
| 379 | for (auto i = 0; i < scoreSize; i++) |
| 380 | { |
| 381 | for (auto j = 0; j < scoreSize; j++) |
| 382 | { |
| 383 | for (auto k = 0; k < anchorNum; k++) |
| 384 | { |
| 385 | anchorIndex[0] = 1, anchorIndex[1] = k, anchorIndex[2] = i, anchorIndex[3] = j; |
| 386 | anchors.at<float>(idx: anchorIndex) = ori + totalStride * i; |
| 387 | |
| 388 | anchorIndex[0] = 0; |
| 389 | anchors.at<float>(idx: anchorIndex) = ori + totalStride * j; |
| 390 | |
| 391 | anchorIndex[0] = 2; |
| 392 | anchors.at<float>(idx: anchorIndex) = baseAnchors[k].width; |
| 393 | |
| 394 | anchorIndex[0] = 3; |
| 395 | anchors.at<float>(idx: anchorIndex) = baseAnchors[k].height; |
| 396 | } |
| 397 | } |
| 398 | } |
| 399 | |
| 400 | return anchors; |
| 401 | } |
| 402 | |
| 403 | Mat TrackerDaSiamRPNImpl::getSubwindow(Mat& img, const Rect2f& targetBox, float originalSize, Scalar avgChans) |
| 404 | { |
| 405 | Mat zCrop, dst; |
| 406 | Size imgSize = img.size(); |
| 407 | float c = (originalSize + 1) / 2; |
| 408 | float xMin = (float)cvRound(value: targetBox.x - c); |
| 409 | float xMax = xMin + originalSize - 1; |
| 410 | float yMin = (float)cvRound(value: targetBox.y - c); |
| 411 | float yMax = yMin + originalSize - 1; |
| 412 | |
| 413 | int leftPad = (int)(fmax(x: 0., y: -xMin)); |
| 414 | int topPad = (int)(fmax(x: 0., y: -yMin)); |
| 415 | int rightPad = (int)(fmax(x: 0., y: xMax - imgSize.width + 1)); |
| 416 | int bottomPad = (int)(fmax(x: 0., y: yMax - imgSize.height + 1)); |
| 417 | |
| 418 | xMin = xMin + leftPad; |
| 419 | xMax = xMax + leftPad; |
| 420 | yMax = yMax + topPad; |
| 421 | yMin = yMin + topPad; |
| 422 | |
| 423 | if (topPad == 0 && bottomPad == 0 && leftPad == 0 && rightPad == 0) |
| 424 | { |
| 425 | img(Rect(int(xMin), int(yMin), int(xMax - xMin + 1), int(yMax - yMin + 1))).copyTo(m: zCrop); |
| 426 | } |
| 427 | else |
| 428 | { |
| 429 | copyMakeBorder(src: img, dst, top: topPad, bottom: bottomPad, left: leftPad, right: rightPad, borderType: BORDER_CONSTANT, value: avgChans); |
| 430 | dst(Rect(int(xMin), int(yMin), int(xMax - xMin + 1), int(yMax - yMin + 1))).copyTo(m: zCrop); |
| 431 | } |
| 432 | |
| 433 | return zCrop; |
| 434 | } |
| 435 | |
| 436 | Ptr<TrackerDaSiamRPN> TrackerDaSiamRPN::create(const TrackerDaSiamRPN::Params& parameters) |
| 437 | { |
| 438 | return makePtr<TrackerDaSiamRPNImpl>(a1: parameters); |
| 439 | } |
| 440 | |
| 441 | Ptr<TrackerDaSiamRPN> TrackerDaSiamRPN::create(const dnn::Net& siam_rpn, const dnn::Net& kernel_cls1, const dnn::Net& kernel_r1) |
| 442 | { |
| 443 | return makePtr<TrackerDaSiamRPNImpl>(a1: siam_rpn, a1: kernel_cls1, a1: kernel_r1); |
| 444 | } |
| 445 | |
| 446 | #else // OPENCV_HAVE_DNN |
| 447 | Ptr<TrackerDaSiamRPN> TrackerDaSiamRPN::create(const TrackerDaSiamRPN::Params& parameters) |
| 448 | { |
| 449 | (void)(parameters); |
| 450 | CV_Error(cv::Error::StsNotImplemented, "to use DaSiamRPN, the tracking module needs to be built with opencv_dnn !" ); |
| 451 | } |
| 452 | #endif // OPENCV_HAVE_DNN |
| 453 | } |
| 454 | |