def draw_bboxes(self, image_canvas, info_tree, conf_threshold=None):
for object in info_tree['objects']:
class_name = self.configer.get('details', 'name_seq')[object['label']]
if 'score' in object:
if conf_threshold is not None and float(object['score']) < conf_threshold:
continue
class_name = '{}_{}'.format(class_name, object['score'])
color_num = len(self.configer.get('details', 'color_list'))
cv2.rectangle(image_canvas,
(int(object['bbox'][0]), int(object['bbox'][1])),
(int(object['bbox'][2]), int(object['bbox'][3])),
color=self.configer.get('details', 'color_list')[object['label'] % color_num], thickness=3)
cv2.putText(image_canvas, class_name,
(int(object['bbox'][0]) + 5, int(object['bbox'][3]) - 5),
cv2.FONT_HERSHEY_SIMPLEX, fontScale=0.5,
color=self.configer.get('details', 'color_list')[object['label'] % color_num], thickness=2)
height, width, _ = image_canvas.shape
if isinstance(object['segm'], list):
maskmap = MaskHelper.polys2mask(object['segm'], width, height)
else:
maskmap = MaskHelper.rle2mask(object['segm'], width, height)
ori_image_canvas = image_canvas.copy()
mask_color = self.configer.get('details', 'color_list')[object['label'] % color_num]
mask_color = np.array(mask_color, dtype=np.uint8)
mask_canvas = np.repeat(maskmap[:, :, np.newaxis], 3, 2).astype(np.uint8) * mask_color
image_canvas = cv2.addWeighted(image_canvas, 0.6, mask_canvas, 0.5, 0)
image_canvas[mask_canvas == 0] = ori_image_canvas[mask_canvas == 0]
return image_canvas
def relabel(self, json_dir, method='mask_rcnn'):
submission_file = os.path.join(
json_dir,
'person_instances_val2017_{}_results.json'.format(method))
img_id_list = list()
object_list = list()
for json_file in os.listdir(json_dir):
json_path = os.path.join(json_dir, json_file)
shotname, extensions = os.path.splitext(json_file)
try:
img_id = int(shotname)
except ValueError:
Log.info('Invalid Json file: {}'.format(json_file))
continue
img_id_list.append(img_id)
with open(json_path, 'r') as json_stream:
info_tree = json.load(json_stream)
for object in info_tree['objects']:
object_dict = dict()
object_dict['image_id'] = img_id
object_dict['category_id'] = int(
self.configer.get('data',
'coco_cat_seq')[object['label']])
object_dict['score'] = object['score']
object_dict['bbox'] = [
object['bbox'][0], object['bbox'][1],
object['bbox'][2] - object['bbox'][0],
object['bbox'][3] - object['bbox'][1]
]
if isinstance(object['segm'], dict):
object_dict['segmentation'] = object['segm']
else:
object_dict['segmentation'] = maskUtils.encode(
np.asfortranarray(
MaskHelper.polys2mask(object['segm'],
info_tree['height'],
info_tree['width'])))
object_list.append(object_dict)
with open(submission_file, 'w') as write_stream:
write_stream.write(json.dumps(object_list))
Log.info('Evaluate {} images...'.format(len(img_id_list)))
return submission_file, img_id_list