Spaces:
Runtime error
Runtime error
| import os | |
| from pathlib import Path | |
| import numpy as np | |
| import cv2 | |
| try: | |
| import detectron2 | |
| except: | |
| os.system('pip install -U pip') | |
| os.system("pip install 'git+https://github.com/facebookresearch/fvcore.git'") | |
| os.system("pip install detectron2 -f https://dl.fbaipublicfiles.com/detectron2/wheels/cu111/torch1.10/index.html") | |
| from detectron2 import model_zoo | |
| from detectron2.engine import DefaultPredictor | |
| from detectron2.config import get_cfg | |
| from detectron2.utils.visualizer import Visualizer | |
| from detectron2.utils.visualizer import ColorMode | |
| PATH_PROJECT = Path(__file__).parent.parent | |
| def get_model(): | |
| """ | |
| This function is for the model of the project | |
| """ | |
| cfg = get_cfg() | |
| cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_50_FPN_3x.yaml")) | |
| cfg.MODEL.ROI_HEADS.SCORE_THRESH_TEST = 0.7 # set threshold for this model | |
| cfg.MODEL.WEIGHTS = str(PATH_PROJECT/"output"/"model_final.pth") # Let training initialize from model zoo | |
| cfg.MODEL.ROI_HEADS.NUM_CLASSES = 2 | |
| predictor = DefaultPredictor(cfg) | |
| return predictor | |
| def predict_image(img_pil): | |
| """ | |
| This function is for the prediction of the model | |
| return the image with the prediction and the areas of the objects | |
| """ | |
| predictor = get_model() | |
| img_array = np.array(img_pil) | |
| outputs = predictor(img_array) | |
| v = Visualizer(img_array, | |
| scale=1, | |
| instance_mode=ColorMode.IMAGE_BW # remove the colors of unsegmented pixels | |
| ) | |
| v = v.draw_instance_predictions(outputs["instances"].to("cpu")) | |
| image_output = v.get_image()[:, :, ::-1].copy() | |
| masks = outputs["instances"].pred_masks.cpu().numpy() | |
| class_ids = outputs["instances"].pred_classes.cpu().numpy() | |
| areas = [np.sum(mask) for mask in masks] | |
| # Add text labels with the object IDs and areas | |
| for i, (mask, class_id, area) in enumerate(zip(masks, class_ids, areas)): | |
| text = f"The id is {i}" | |
| text_size, _ = cv2.getTextSize(text, cv2.FONT_HERSHEY_SIMPLEX, fontScale=0.5, thickness=1) | |
| pos = (np.unravel_index(np.argmax(mask), mask.shape))[::-1] | |
| pos = (pos[0] - text_size[0]//2, pos[1] - text_size[1]//2) | |
| cv2.putText(image_output, text, pos, cv2.FONT_HERSHEY_SIMPLEX, fontScale=0.5, color=(0,0,255), thickness=1) | |
| values = {"image":image_output, "areas":areas, "masks":masks} | |
| return values | |