def main():
cfg = setup_cfg()
demo = VisualizationDemo(cfg)
args = get_parser().parse_args()
img_path = args.img_path
new_img_path = args.new_img_path
save_path = args.save_path
for index, path in enumerate(tqdm(os.listdir(img_path))):
if path.endswith('.jpg'):
fullpath = os.path.join(img_path, path)
img = read_image(fullpath, format='BGR')
# remove the text 'google' from image
img = img[:620, :, :]
pred, out, poly = demo.run_on_image(img)
if poly == []:
cv2.imwrite(os.path.join(save_path, path), img,
[cv2.IMWRITE_JPEG_QUALITY, 75])
os.remove(fullpath)
else:
cv2.imwrite(os.path.join(new_img_path, path), img,
[cv2.IMWRITE_JPEG_QUALITY, 75])
os.remove(fullpath)
def predict_API():
'''
供外部调用，返回一个初始化完成的模型，可以直接用来预测。搭建后台服务器用来响应前端页面的时候会用到。
:return:
'''
args = get_parser().parse_args()
cfg = setup_cfg(args)
register_dataset()
demo = VisualizationDemo(cfg, instance_mode=ColorMode.SEGMENTATION)
return demo
def do_crop():
args = get_parser().parse_args()
logger = setup_logger()
logger.info("Arguments: " + str(args))
cfg = setup_cfg(args)
# 注册数据集
register_dataset()
model_demo = VisualizationDemo(cfg)
face_img_path = 'face_img/'
all_img_path = 'all_img_json/'
for index, imgfile in enumerate(os.listdir(all_img_path)):
# use PIL, to be consistent with evaluation
img_fullName = os.path.join(all_img_path, imgfile)
img = read_image(img_fullName, format="BGR")
start_time = time.time()
predictions = model_demo.predictor(img)
bbox = model_demo.highest_only(predictions).get('pred_boxes')
if len(bbox.tensor) == 0:
print(imgfile)
continue
bbox = bbox.tensor[0].tolist()
all_label = tools.get_all_class_label_infos()
json_template = tools.labelme_template()
json_template["imageHeight"], json_template["imageWidth"] = img.shape[:2]
json_template["imagePath"] = imgfile
json_template["imageData"] = str(img_base64(img_fullName), encoding="utf-8")
json_template["shapes"][0]["label"] = all_label[imgfile]
json_template["shapes"][0]["points"] = [[bbox[0], bbox[1]], [bbox[2], bbox[3]]]
json.dump(json_template, open(os.path.join(all_img_path, imgfile.replace('.jpg', '.json')), 'w'), indent=1)
# img = Image.open(img_fullName)
# img = img.crop((bbox[0], bbox[1], bbox[2], bbox[3]))
# img.save(os.path.join(face_img_path, imgfile))
if index % 1000 == 0:
print(index)
default=[],
nargs=argparse.REMAINDER,
)
return parser
if __name__ == "__main__":
mp.set_start_method("spawn", force=True)
args = get_parser().parse_args()
setup_logger(name="fvcore")
logger = setup_logger()
logger.info("Arguments: " + str(args))
cfg = setup_cfg(args)
demo = VisualizationDemo(cfg)
if args.input:
if len(args.input) == 1:
args.input = glob.glob(os.path.expanduser(args.input[0]))
assert args.input, "The input path(s) was not found"
for path in tqdm.tqdm(args.input, disable=not args.output):
# use PIL, to be consistent with evaluation
img = read_image(path, format="BGR")
start_time = time.time()
predictions, visualized_output = demo.run_on_image(
img, args.confidence_threshold)
logger.info("{}: {} in {:.2f}s".format(
path,
"detected {} instances".format(len(predictions["instances"]))
if "instances" in predictions else "finished",
default=[],
nargs=argparse.REMAINDER,
)
return parser
if __name__ == "__main__":
mp.set_start_method("spawn", force=True)
args = get_parser().parse_args()
setup_logger(name="fvcore")
logger = setup_logger()
logger.info("Arguments: " + str(args))
cfg = setup_cfg(args)
demo = VisualizationDemo(cfg)
if args.input:
if len(args.input) == 1:
args.input = glob.glob(os.path.expanduser(args.input[0]))
assert args.input, "The input path(s) was not found"
for path in tqdm.tqdm(args.input, disable=not args.output):
# use PIL, to be consistent with evaluation
img = read_image(path, format="BGR")
start_time = time.time()
predictions, visualized_output = demo.run_on_image(img)
logger.info("{}: {} in {:.2f}s".format(
path,
"detected {} instances".format(len(predictions["instances"]))
if "instances" in predictions else "finished",
time.time() - start_time,
return predictions
if __name__ == "__main__":
ALL_START = time.time()
print("Start counting time. ALL_START")
# mp.set_start_method("spawn", force=True)
args = get_parser().parse_args()
setup_logger(name="fvcore")
logger = setup_logger()
logger.info("Arguments: " + str(args))
if args.config_file.endswith("yaml"):
cfg = setup_cfg(args)
model = VisualizationDemo(cfg)
elif args.config_file.endswith("py"):
model = model_zoo.get(args.config_file, trained=True)
model.cuda()
model.eval()
else:
raise Exception(f"{args.config_file} is not supported.")
if args.input:
if len(args.input) == 1:
args.input = glob.glob(osp.expanduser(args.input[0]))
assert args.input, "The input path(s) was not found"
data_dir = osp.join(args.input[0], args.split)
if args.video_src_names: # only process the given video sources
video_src_names = args.video_src_names
from demo.predictor import VisualizationDemo
with open('data/objects_vocab.txt', 'r') as f:
obj_vocab = f.read().splitlines()
# Registering metadata for Visual Genome Object Classes
# MetadataCatalog.get("VisualGenomeObjects").thing_classes = obj_vocab
cfg = get_cfg()
cfg.merge_from_file(
'configs/COCO-Detection/faster_rcnn_X_101_64x4d_FPN_2x_vlp.yaml')
cfg.freeze()
predictor = DefaultPredictor(cfg)
demo = VisualizationDemo(cfg)
#model = build_model(cfg)
#model.eval()
#checkpointer = DetectionCheckpointer(model)
#checkpointer.load(os.path.join('model_weights', 'e2e_faster_rcnn_X-101-64x4d-FPN_2x-vlp.pkl'))
img1 = read_image(os.path.join('test_images', '12283150_12d37e6389_z.jpg'),
format="BGR")
img2 = read_image(os.path.join('test_images', '25691390_f9944f61b5_z.jpg'),
format="BGR")
img3 = read_image(os.path.join('test_images', '9247489789_132c0d534a_z.jpg'),
format="BGR")
preds, vis_out = demo.run_on_image(img3, obj_vocab)
)
return parser
if __name__ == "__main__":
mp.set_start_method("spawn", force=True)
args = get_parser().parse_args()
logger = setup_logger()
logger.info("Arguments: " + str(args))
cfg = setup_cfg(args)
# 注册数据集
register_dataset()
demo = VisualizationDemo(cfg, instance_mode=ColorMode.SEGMENTATION)
# for path in tqdm.tqdm(args.input, disable=not args.output):
for imgfile in os.listdir(INPUT_IMG_PATH):
# use PIL, to be consistent with evaluation
img_fullName = os.path.join(INPUT_IMG_PATH, imgfile)
img = read_image(img_fullName, format="BGR")
start_time = time.time()
predictions, visualized_output = demo.run_on_image(img)
logger.info("{}: detected {} instances in {:.2f}s".format(
imgfile, len(predictions["instances"]),
time.time() - start_time))
if args.output:
if os.path.isdir(args.output):
"is_thresh"]
cfg.freeze()
return cfg
if __name__ == "__main__":
yaml_path = os.path.join(
Path(__file__).parent.parent, "params", "azure_centermask_SNU.yaml")
with open(yaml_path) as f:
params = yaml.load(f, Loader=yaml.FullLoader)
cfg = setup_cfg(params)
os.environ["CUDA_VISIBLE_DEVICES"] = params["is_gpu_id"]
device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
print("==> Loading CenterMask on", device, params["is_gpu_id"])
demo = VisualizationDemo(cfg)
sock = su.initialize_client(params["tcp_ip"],
params["centermask_tcp_port"])
while True:
img = su.recvall_image(sock)
predictions, vis_output = demo.run_on_image(img)
pred_masks = predictions["instances"].pred_masks.cpu().detach().numpy(
) # (N, H, W),
pred_boxes = predictions["instances"].pred_boxes.tensor.cpu().detach(
).numpy() # (x1, y1, x2, y2)
pred_scores = predictions["instances"].scores.cpu().detach().numpy(
) # a vector of N confidence scores.
pred_classes = predictions["instances"].pred_classes.cpu().detach(
).numpy() # [0, num_categories).
vis_img = cv2.resize(vis_output.get_image()[:, :, ::-1],
cfg.merge_from_file(model_zoo.get_config_file("COCO-PanopticSegmentation/panoptic_fpn_R_101_3x.yaml"))
cfg.MODEL.WEIGHTS = ROOT_DIR + "/configs/COCO-PanopticSegmentation/panoptic_fpn_R_101_3x_model.pkl"
elif segmentation == 'mrcnn':
cfg.merge_from_file(model_zoo.get_config_file("COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x.yaml"))
cfg.MODEL.WEIGHTS = '/home/users/t/tim.schroeder/object_rep/project_code/detectron2/configs/COCO-InstanceSegmentation/mask_rcnn_R_101_FPN_3x_model.pkl'
if not hpc:
cfg.MODEL.DEVICE = 'cpu' #only on local device
cfg.freeze()
return cfg
if __name__ == "__main__":
cfg = setup_cfg()
predictor = DefaultPredictor(cfg)
demo = VisualizationDemo(cfg)
video_path_list = [video for video in glob.glob(DATA_DIR + '/*')]
video_path_list.sort()
for video_path in [video_path_list[12]]:
output_data = []
output_info = []
video = cv2.VideoCapture(video_path)
width = int(video.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(video.get(cv2.CAP_PROP_FRAME_HEIGHT))
frames_per_second = video.get(cv2.CAP_PROP_FPS)
num_frames = int(video.get(cv2.CAP_PROP_FRAME_COUNT))
basename = basename = os.path.splitext(os.path.basename(video_path))[0]
if os.path.isdir(RESULTS_DIR):
def matchScore(x, y, mask):
center = center_of_mass(mask)
distanceFromCenter = euclidean(center, (y, x))
size = np.count_nonzero(mask)
maxDist = 1468
maxSize = 921600
score = (((1 / (maxSize**2)) * (size - maxSize)**2) +
(1 - distanceFromCenter / maxDist)) / 2
return score
if __name__ == '__main__':
cfg = setup_cfg() #Define Classifier for Panoptic Segmentation
predictor = DefaultPredictor(cfg)
demo = VisualizationDemo(cfg)
metadata = demo.metadata
GAZE_FOLDER_DIR = '/mnt/raid/data/SCIoI/GazeCom_ground_truth/'
gaze_folder_list = [video for video in glob.glob(GAZE_FOLDER_DIR + '/*')]
gaze_folder_list.sort()
VIDEO_DIR = '/mnt/raid/data/SCIoI/GazeCom_videos/'
video_path_list = [video for video in glob.glob(VIDEO_DIR + '/*')]
video_path_list.sort()
PROTO_OBJECTS = '/mnt/raid/data/SCIoI/tim/salData/'
detect_path_list = [
video for video in glob.glob(RESULTS_DIR + 'detect_video/' + '*.avi')
]
def do_test():
mp.set_start_method("spawn", force=True)
args = get_parser().parse_args()
logger = setup_logger()
logger.info("Arguments: " + str(args))
cfg = setup_cfg(args)
# 注册数据集
register_dataset()
model_demo = VisualizationDemo(cfg)
predict_list = []
real_list = []
# for path in tqdm.tqdm(args.input, disable=not args.output):
for imgfile in os.listdir(INPUT_IMG_PATH):
# use PIL, to be consistent with evaluation
img_fullName = os.path.join(INPUT_IMG_PATH, imgfile)
img = read_image(img_fullName, format="BGR")
start_time = time.time()
predictions, visualized_output = model_demo.run_on_image_detection(img)
# imgpath.json这个文件里面是图片文件名和类别真值的对应表
# 模型输出的类别中 0和1 分别代表水印和正常元件，而 2，3，4，5，6 分别是那五种瑕疵，
# 还缺少一个表示false的数字，所以我们用7来表示
# 所以在imgpath.json文件中你只能看到2,3,4,5,6,7这6种数字，其实分别就代表了我们最终要分类的6个类别
img_paths = tools.get_all_class_label_infos()
pred = predictions.get('pred_classes').tolist()
if len(pred) == 0:
print(imgfile)
predict_list.append('1')
else:
predict_list.append(str(pred[0] + 1))
# temp = str(imgfile).split('_')
# if temp[1].endswith('.jpg'):
# imgfile = temp[0] + '_' + temp[1]
# else:
# imgfile = temp[0] + '_' + temp[1] + '.jpg'
real_list.append(img_paths[imgfile])
#visualized_output.save(os.path.join('imgout', imgfile))
# print(flaw_only(predictions))
# log会在控制台输出预测所花的时间等等信息
# logger.info(
# "{}: detected {} instances in {:.2f}s".format(
# imgfile, len(predictions["instances"]), time.time() - start_time
# )
# )
# 这个是保存预测错误的图片，会在原图上额外绘画出识别的矩形框、mask、类别、分数等等，好让我们看看预测错误的图片长什么样并且是哪里预测错了
# if determine_final_class(predictions) != img_paths[imgfile]:
# temp = ['watermark', 'commom1', '01', '02', '05', '07', 'other', 'false']
# tempstr = temp[img_paths[imgfile]] + '-' + temp[determine_final_class(predictions)]
# out_filename = os.path.join(OUTPUT_IMG_PATH, tempstr + imgfile)
# visualized_output.save(out_filename)
# 根据上面两种函数，determine_final_class和contain_class两种方案来统计的情况，生成混淆矩阵、precision、recall
cm = confusion_matrix(real_list, predict_list)
ps = precision_score(real_list, predict_list, average=None)
rc = recall_score(real_list, predict_list, average=None)
print(cm)
print(['{:.2%}'.format(x) for x in ps])
print(['{:.2%}'.format(x) for x in rc])
'''