def detect_face(self, img): """Detect face over image""" boxes_align = torch.Tensor([]) landmark_align = torch.Tensor([]) img = image_tools.convert_image_to_tensor(img).unsqueeze(0) if self.use_cuda: img = img.cuda() # pnet if self.pnet_detector: boxes, boxes_align = self.detect_pnet(img.clone()) if boxes_align is None: return torch.Tensor([]), torch.Tensor([]) # rnet if self.rnet_detector: boxes, boxes_align = self.detect_rnet(img.clone(), boxes_align) if boxes_align is None: return torch.Tensor([]), torch.Tensor([]) # onet if self.onet_detector: boxes_align, landmark_align = self.detect_onet( img.clone(), boxes_align) if boxes_align is None: return torch.Tensor([]), torch.Tensor([]) return boxes_align, landmark_align
def detect_onet(self, im, dets): """Get face candidates using onet Parameters: ---------- im: numpy array input image array dets: numpy array detection results of rnet Returns: ------- boxes_align: numpy array boxes after calibration landmarks_align: numpy array landmarks after calibration """ h, w, c = im.shape if dets is None: return None, None dets = self.square_bbox(dets) dets[:, 0:4] = np.round(dets[:, 0:4]) [dy, edy, dx, edx, y, ey, x, ex, tmpw, tmph] = self.pad(dets, w, h) num_boxes = dets.shape[0] # cropped_ims_tensors = np.zeros((num_boxes, 3, 24, 24), dtype=np.float32) cropped_ims_tensors = [] for i in range(num_boxes): tmp = np.zeros((tmph[i], tmpw[i], 3), dtype=np.uint8) tmp[dy[i]:edy[i] + 1, dx[i]:edx[i] + 1, :] = im[y[i]:ey[i] + 1, x[i]:ex[i] + 1, :] crop_im = cv2.resize(tmp, (48, 48)) crop_im_tensor = image_tools.convert_image_to_tensor(crop_im) # cropped_ims_tensors[i, :, :, :] = crop_im_tensor cropped_ims_tensors.append(crop_im_tensor) feed_imgs = Variable(torch.stack(cropped_ims_tensors)) if self.rnet_detector.use_cuda: feed_imgs = feed_imgs.cuda() cls_map, reg, landmark = self.onet_detector(feed_imgs) cls_map = cls_map.cpu().data.numpy() reg = reg.cpu().data.numpy() landmark = landmark.cpu().data.numpy() keep_inds = np.where(cls_map > self.thresh[2])[0] if len(keep_inds) > 0: boxes = dets[keep_inds] cls = cls_map[keep_inds] reg = reg[keep_inds] landmark = landmark[keep_inds] else: return None, None keep = utils.nms(boxes, 0.7, mode="Minimum") if len(keep) == 0: return None, None keep_cls = cls[keep] keep_boxes = boxes[keep] keep_reg = reg[keep] keep_landmark = landmark[keep] bw = keep_boxes[:, 2] - keep_boxes[:, 0] + 1 bh = keep_boxes[:, 3] - keep_boxes[:, 1] + 1 align_topx = keep_boxes[:, 0] + keep_reg[:, 0] * bw align_topy = keep_boxes[:, 1] + keep_reg[:, 1] * bh align_bottomx = keep_boxes[:, 2] + keep_reg[:, 2] * bw align_bottomy = keep_boxes[:, 3] + keep_reg[:, 3] * bh align_landmark_topx = keep_boxes[:, 0] align_landmark_topy = keep_boxes[:, 1] boxes_align = np.vstack([ align_topx, align_topy, align_bottomx, align_bottomy, keep_cls[:, 0], # align_topx + keep_landmark[:, 0] * bw, # align_topy + keep_landmark[:, 1] * bh, # align_topx + keep_landmark[:, 2] * bw, # align_topy + keep_landmark[:, 3] * bh, # align_topx + keep_landmark[:, 4] * bw, # align_topy + keep_landmark[:, 5] * bh, # align_topx + keep_landmark[:, 6] * bw, # align_topy + keep_landmark[:, 7] * bh, # align_topx + keep_landmark[:, 8] * bw, # align_topy + keep_landmark[:, 9] * bh, ]) boxes_align = boxes_align.T landmark = np.vstack([ align_landmark_topx + keep_landmark[:, 0] * bw, align_landmark_topy + keep_landmark[:, 1] * bh, align_landmark_topx + keep_landmark[:, 2] * bw, align_landmark_topy + keep_landmark[:, 3] * bh, align_landmark_topx + keep_landmark[:, 4] * bw, align_landmark_topy + keep_landmark[:, 5] * bh, align_landmark_topx + keep_landmark[:, 6] * bw, align_landmark_topy + keep_landmark[:, 7] * bh, align_landmark_topx + keep_landmark[:, 8] * bw, align_landmark_topy + keep_landmark[:, 9] * bh, ]) landmark_align = landmark.T return boxes_align, landmark_align
def detect_rnet(self, im, dets): """Get face candidates using rnet Parameters: ---------- im: numpy array input image array dets: numpy array detection results of pnet Returns: ------- boxes: numpy array detected boxes before calibration boxes_align: numpy array boxes after calibration """ h, w, c = im.shape if dets is None: return None, None dets = self.square_bbox(dets) dets[:, 0:4] = np.round(dets[:, 0:4]) [dy, edy, dx, edx, y, ey, x, ex, tmpw, tmph] = self.pad(dets, w, h) num_boxes = dets.shape[0] ''' # helper for setting RNet batch size batch_size = self.rnet_detector.batch_size ratio = float(num_boxes) / batch_size if ratio > 3 or ratio < 0.3: print "You may need to reset RNet batch size if this info appears frequently, \ face candidates:%d, current batch_size:%d"%(num_boxes, batch_size) ''' # cropped_ims_tensors = np.zeros((num_boxes, 3, 24, 24), dtype=np.float32) cropped_ims_tensors = [] for i in range(num_boxes): tmp = np.zeros((tmph[i], tmpw[i], 3), dtype=np.uint8) tmp[dy[i]:edy[i] + 1, dx[i]:edx[i] + 1, :] = im[y[i]:ey[i] + 1, x[i]:ex[i] + 1, :] crop_im = cv2.resize(tmp, (24, 24)) crop_im_tensor = image_tools.convert_image_to_tensor(crop_im) # cropped_ims_tensors[i, :, :, :] = crop_im_tensor cropped_ims_tensors.append(crop_im_tensor) feed_imgs = Variable(torch.stack(cropped_ims_tensors)) if self.rnet_detector.use_cuda: feed_imgs = feed_imgs.cuda() cls_map, reg = self.rnet_detector(feed_imgs) cls_map = cls_map.cpu().data.numpy() reg = reg.cpu().data.numpy() # landmark = landmark.cpu().data.numpy() keep_inds = np.where(cls_map > self.thresh[1])[0] if len(keep_inds) > 0: boxes = dets[keep_inds] cls = cls_map[keep_inds] reg = reg[keep_inds] # landmark = landmark[keep_inds] else: return None, None keep = utils.nms(boxes, 0.7) if len(keep) == 0: return None, None keep_cls = cls[keep] keep_boxes = boxes[keep] keep_reg = reg[keep] # keep_landmark = landmark[keep] bw = keep_boxes[:, 2] - keep_boxes[:, 0] + 1 bh = keep_boxes[:, 3] - keep_boxes[:, 1] + 1 boxes = np.vstack([ keep_boxes[:, 0], keep_boxes[:, 1], keep_boxes[:, 2], keep_boxes[:, 3], keep_cls[:, 0], # keep_boxes[:,0] + keep_landmark[:, 0] * bw, # keep_boxes[:,1] + keep_landmark[:, 1] * bh, # keep_boxes[:,0] + keep_landmark[:, 2] * bw, # keep_boxes[:,1] + keep_landmark[:, 3] * bh, # keep_boxes[:,0] + keep_landmark[:, 4] * bw, # keep_boxes[:,1] + keep_landmark[:, 5] * bh, # keep_boxes[:,0] + keep_landmark[:, 6] * bw, # keep_boxes[:,1] + keep_landmark[:, 7] * bh, # keep_boxes[:,0] + keep_landmark[:, 8] * bw, # keep_boxes[:,1] + keep_landmark[:, 9] * bh, ]) align_topx = keep_boxes[:, 0] + keep_reg[:, 0] * bw align_topy = keep_boxes[:, 1] + keep_reg[:, 1] * bh align_bottomx = keep_boxes[:, 2] + keep_reg[:, 2] * bw align_bottomy = keep_boxes[:, 3] + keep_reg[:, 3] * bh boxes_align = np.vstack([ align_topx, align_topy, align_bottomx, align_bottomy, keep_cls[:, 0], # align_topx + keep_landmark[:, 0] * bw, # align_topy + keep_landmark[:, 1] * bh, # align_topx + keep_landmark[:, 2] * bw, # align_topy + keep_landmark[:, 3] * bh, # align_topx + keep_landmark[:, 4] * bw, # align_topy + keep_landmark[:, 5] * bh, # align_topx + keep_landmark[:, 6] * bw, # align_topy + keep_landmark[:, 7] * bh, # align_topx + keep_landmark[:, 8] * bw, # align_topy + keep_landmark[:, 9] * bh, ]) boxes = boxes.T boxes_align = boxes_align.T return boxes, boxes_align
def detect_pnet(self, im): """Get face candidates through pnet Parameters: ---------- im: numpy array input image array Returns: ------- boxes: numpy array detected boxes before calibration boxes_align: numpy array boxes after calibration """ # im = self.unique_image_format(im) h, w, c = im.shape net_size = 12 current_scale = float( net_size) / self.min_face_size # find initial scale im_resized = self.resize_image(im, current_scale) current_height, current_width, _ = im_resized.shape # fcn all_boxes = list() while min(current_height, current_width) > net_size: feed_imgs = [] image_tensor = image_tools.convert_image_to_tensor(im_resized) feed_imgs.append(image_tensor) feed_imgs = torch.stack(feed_imgs) feed_imgs = Variable(feed_imgs) if self.pnet_detector.use_cuda: feed_imgs = feed_imgs.cuda() cls_map, reg = self.pnet_detector(feed_imgs) cls_map_np = image_tools.convert_chwTensor_to_hwcNumpy( cls_map.cpu()) reg_np = image_tools.convert_chwTensor_to_hwcNumpy(reg.cpu()) # landmark_np = image_tools.convert_chwTensor_to_hwcNumpy(landmark.cpu()) boxes = self.generate_bounding_box(cls_map_np[0, :, :], reg_np, current_scale, self.thresh[0]) current_scale *= self.scale_factor im_resized = self.resize_image(im, current_scale) current_height, current_width, _ = im_resized.shape if boxes.size == 0: continue keep = utils.nms(boxes[:, :5], 0.5, 'Union') boxes = boxes[keep] all_boxes.append(boxes) if len(all_boxes) == 0: return None, None all_boxes = np.vstack(all_boxes) # merge the detection from first stage keep = utils.nms(all_boxes[:, 0:5], 0.7, 'Union') all_boxes = all_boxes[keep] # boxes = all_boxes[:, :5] bw = all_boxes[:, 2] - all_boxes[:, 0] + 1 bh = all_boxes[:, 3] - all_boxes[:, 1] + 1 # landmark_keep = all_boxes[:, 9:].reshape((5,2)) boxes = np.vstack([ all_boxes[:, 0], all_boxes[:, 1], all_boxes[:, 2], all_boxes[:, 3], all_boxes[:, 4], # all_boxes[:, 0] + all_boxes[:, 9] * bw, # all_boxes[:, 1] + all_boxes[:,10] * bh, # all_boxes[:, 0] + all_boxes[:, 11] * bw, # all_boxes[:, 1] + all_boxes[:, 12] * bh, # all_boxes[:, 0] + all_boxes[:, 13] * bw, # all_boxes[:, 1] + all_boxes[:, 14] * bh, # all_boxes[:, 0] + all_boxes[:, 15] * bw, # all_boxes[:, 1] + all_boxes[:, 16] * bh, # all_boxes[:, 0] + all_boxes[:, 17] * bw, # all_boxes[:, 1] + all_boxes[:, 18] * bh ]) boxes = boxes.T align_topx = all_boxes[:, 0] + all_boxes[:, 5] * bw align_topy = all_boxes[:, 1] + all_boxes[:, 6] * bh align_bottomx = all_boxes[:, 2] + all_boxes[:, 7] * bw align_bottomy = all_boxes[:, 3] + all_boxes[:, 8] * bh # refine the boxes boxes_align = np.vstack([ align_topx, align_topy, align_bottomx, align_bottomy, all_boxes[:, 4], # align_topx + all_boxes[:,9] * bw, # align_topy + all_boxes[:,10] * bh, # align_topx + all_boxes[:,11] * bw, # align_topy + all_boxes[:,12] * bh, # align_topx + all_boxes[:,13] * bw, # align_topy + all_boxes[:,14] * bh, # align_topx + all_boxes[:,15] * bw, # align_topy + all_boxes[:,16] * bh, # align_topx + all_boxes[:,17] * bw, # align_topy + all_boxes[:,18] * bh, ]) boxes_align = boxes_align.T return boxes, boxes_align
def train_pnet(model_store_path, end_epoch, imdb, batch_size, frequent=50, base_lr=0.01, use_cuda=True): if not os.path.exists(model_store_path): os.makedirs(model_store_path) lossfn = LossFn() net = PNet(is_train=True, use_cuda=use_cuda) net.train() if use_cuda: net.cuda() optimizer = torch.optim.Adam(net.parameters(), lr=base_lr) train_data = TrainImageReader(imdb, 12, batch_size, shuffle=True) for cur_epoch in range(1, end_epoch + 1): train_data.reset() accuracy_list = [] cls_loss_list = [] bbox_loss_list = [] # landmark_loss_list=[] for batch_idx, (image, (gt_label, gt_bbox, gt_landmark)) in enumerate(train_data): im_tensor = [ image_tools.convert_image_to_tensor(image[i, :, :, :]) for i in range(image.shape[0]) ] im_tensor = torch.stack(im_tensor) im_tensor = Variable(im_tensor) gt_label = Variable(torch.from_numpy(gt_label).float()) gt_bbox = Variable(torch.from_numpy(gt_bbox).float()) # gt_landmark = Variable(torch.from_numpy(gt_landmark).float()) if use_cuda: im_tensor = im_tensor.cuda() gt_label = gt_label.cuda() gt_bbox = gt_bbox.cuda() # gt_landmark = gt_landmark.cuda() cls_pred, box_offset_pred = net(im_tensor) # all_loss, cls_loss, offset_loss = lossfn.loss(gt_label=label_y,gt_offset=bbox_y, pred_label=cls_pred, pred_offset=box_offset_pred) cls_loss = lossfn.cls_loss(gt_label, cls_pred) box_offset_loss = lossfn.box_loss(gt_label, gt_bbox, box_offset_pred) # landmark_loss = lossfn.landmark_loss(gt_label,gt_landmark,landmark_offset_pred) all_loss = cls_loss * 1.0 + box_offset_loss * 0.5 if batch_idx % frequent == 0: accuracy = compute_accuracy(cls_pred, gt_label) show1 = accuracy.data.tolist() show2 = cls_loss.data.tolist() show3 = box_offset_loss.data.tolist() show5 = all_loss.data.tolist() print( "%s : Epoch: %d, Step: %d, accuracy: %s, det loss: %s, bbox loss: %s, all_loss: %s, lr:%s " % (datetime.datetime.now(), cur_epoch, batch_idx, show1, show2, show3, show5, base_lr)) accuracy_list.append(accuracy) cls_loss_list.append(cls_loss) bbox_loss_list.append(box_offset_loss) optimizer.zero_grad() all_loss.backward() optimizer.step() accuracy_avg = torch.mean(torch.stack(accuracy_list)) cls_loss_avg = torch.mean(torch.stack(cls_loss_list)) bbox_loss_avg = torch.mean(torch.stack(bbox_loss_list)) # landmark_loss_avg = torch.mean(torch.stack(landmark_loss_list)) show6 = accuracy_avg.data.tolist() show7 = cls_loss_avg.data.tolist() show8 = bbox_loss_avg.data.tolist() print("Epoch: %d, accuracy: %s, cls loss: %s, bbox loss: %s" % (cur_epoch, show6, show7, show8)) torch.save( net.state_dict(), os.path.join(model_store_path, "pnet_epoch_%d.pt" % cur_epoch)) torch.save( net, os.path.join(model_store_path, "pnet_epoch_model_%d.pkl" % cur_epoch))
def train_pnet(model_store_path, end_epoch,imdb, batch_size,frequent=50,base_lr=0.01,use_cuda=True): if not os.path.exists(model_store_path): os.makedirs(model_store_path) lossfn = LossFn() net = PNet(is_train=True, use_cuda=use_cuda) net.train() if use_cuda: net.cuda() net = torch.nn.DataParallel(net, device_ids=range(torch.cuda.device_count())) print(net) optimizer = torch.optim.Adam(net.parameters(), lr=base_lr) scheduler = torch.optim.lr_scheduler.MultiStepLR(optimizer, milestones=[10, 25, 40, 45], gamma=0.1) # define the binarization operator bin_op = bin_util.BinOp(net) train_data=TrainImageReader(imdb,12,batch_size,shuffle=True) accuracy_avg_list = [] cls_loss_avg_list = [] bbox_loss_avg_list = [] all_loss_avg_list = [] x1 = range(0, 50) x2 = range(0, 50) x3 = range(0, 50) x4 = range(0, 50) for cur_epoch in range(0, end_epoch): scheduler.step() train_data.reset() accuracy_list=[] cls_loss_list=[] bbox_loss_list=[] # landmark_loss_list=[] all_loss_list = [] for batch_idx,(image,(gt_label,gt_bbox,gt_landmark))in enumerate(train_data): im_tensor = [ image_tools.convert_image_to_tensor(image[i,:,:,:]) for i in range(image.shape[0]) ] im_tensor = torch.stack(im_tensor) im_tensor = Variable(im_tensor) gt_label = Variable(torch.from_numpy(gt_label).float()) gt_bbox = Variable(torch.from_numpy(gt_bbox).float()) # gt_landmark = Variable(torch.from_numpy(gt_landmark).float()) # !!!权重(参数)二值化 !!! bin_op.binarization() #含缩放因子 if use_cuda: im_tensor = im_tensor.cuda() gt_label = gt_label.cuda() gt_bbox = gt_bbox.cuda() # gt_landmark = gt_landmark.cuda() cls_pred, box_offset_pred = net(im_tensor)#含缩放因子 # all_loss, cls_loss, offset_loss = lossfn.loss(gt_label=label_y,gt_offset=bbox_y, pred_label=cls_pred, pred_offset=box_offset_pred) cls_loss = lossfn.cls_loss(gt_label,cls_pred) box_offset_loss = lossfn.box_loss(gt_label,gt_bbox,box_offset_pred) # landmark_loss = lossfn.landmark_loss(gt_label,gt_landmark,landmark_offset_pred) all_loss = cls_loss*1.0+box_offset_loss*0.5 if batch_idx%frequent==0: accuracy=compute_accuracy(cls_pred,gt_label) show1 = str(accuracy.data.tolist()) show2 = str(cls_loss.data.tolist()) show3 = str(box_offset_loss.data.tolist()) show5 = str(all_loss.data.tolist()) print("%s : Epoch: %d, Step: %d, accuracy: %s, cls loss: %s, bbox loss: %s, all_loss: %s, lr:%.6f "%(datetime.datetime.now(),cur_epoch,batch_idx, show1,show2,show3,show5,scheduler.get_lr()[0])) accuracy_list.append(accuracy) cls_loss_list.append(cls_loss) bbox_loss_list.append(box_offset_loss) all_loss_list.append(all_loss) optimizer.zero_grad() all_loss.backward() #计算梯度(指定loss) bin_op.restore() bin_op.updateBinaryGradWeight() #加入缩放因子 optimizer.step() #使用梯度,更新参数(指定optimizer) accuracy_avg = torch.mean(torch.stack(accuracy_list, dim=0)) accuracy_avg_list.append(accuracy_avg) cls_loss_avg = torch.mean(torch.stack(cls_loss_list, dim=0)) cls_loss_avg_list.append(cls_loss_avg) bbox_loss_avg = torch.mean(torch.stack(bbox_loss_list, dim=0)) bbox_loss_avg_list.append(bbox_loss_avg) # landmark_loss_avg = torch.mean(torch.cat(landmark_loss_list)) all_loss_avg = torch.mean(torch.stack(all_loss_list, dim=0)) all_loss_avg_list.append(all_loss_avg) show6 = str(accuracy_avg.data.tolist()) show7 = str(cls_loss_avg.data.tolist()) show8 = str(bbox_loss_avg.data.tolist()) show10 = str(all_loss_avg.data.tolist()) print("Epoch: %d, accuracy: %s, cls loss: %s, bbox loss: %s, all_loss: %s" % (cur_epoch, show6, show7, show8, show10)) #net = net.module torch.save(net.module.state_dict(), os.path.join(model_store_path,"3_bin_pnet_epoch_%d.pt" % cur_epoch)) torch.save(net.module, os.path.join(model_store_path,"3_bin_pnet_epoch_model_%d.pkl" % cur_epoch)) y1 = accuracy_avg_list y2 = cls_loss_avg_list y3 = bbox_loss_avg_list y4 = all_loss_avg_list plt.subplot(1, 4, 1) plt.title('Bin-P-Net') plt.plot(x1, y1, 'o-') plt.xlabel('Epoches') plt.ylabel('Accuracy') plt.subplot(1, 4, 2) plt.plot(x2, y2, 'o-') plt.xlabel('Epoches') plt.ylabel('Cls_loss') plt.subplot(1, 4, 3) plt.plot(x3, y3, 'o-') plt.xlabel('Epoches') plt.ylabel('Bbox_loss') plt.subplot(1, 4, 4) plt.plot(x4, y4, 'o-') plt.xlabel('Epoches') plt.ylabel('All_loss') plt.show() plt.savefig("Bin-accuracy-epoches.jpg")