def ap_for_image(cfg, checkpoint, image_dir, anno_dir, out_dir):
model = init_detector(cfg, checkpoint, device='cuda:0')
images = os.listdir(image_dir)
anno_files = os.listdir(anno_dir)
output_name = 'precision.json'
if not os.path.exists(out_dir):
os.makedirs(out_dir)
output_name = os.path.join(out_dir, output_name)
for image, anno in zip(images, anno_files):
assert image.split('.')[0] == anno.split(
'.')[0], "image and annotation mismatch!"
image_path = os.path.join(image_dir, image)
json_name = os.path.join(anno_dir, anno)
results = inference_detector(model, image_path)
annotations = json_to_annotation(json_name)
mean_ap, _ = eval_map(results[0], annotations)
with open(output_name, 'w') as f:
json.dump([annotations['image_id'], mean_ap], f)
def bbox_map_eval(self, det_result, annotation):
# use only bbox det result
if isinstance(det_result, tuple):
bbox_det_result = [det_result[0]]
else:
bbox_det_result = [det_result]
# mAP
iou_thrs = np.linspace(.5,
0.95,
int(np.round((0.95 - .5) / .05)) + 1,
endpoint=True)
mean_aps = []
for thr in iou_thrs:
mean_ap, _ = eval_map(bbox_det_result, [annotation],
iou_thr=thr,
logger='silent')
mean_aps.append(mean_ap)
return sum(mean_aps) / len(mean_aps)
def bbox_map_eval(det_result, annotation):
"""Evaluate mAP of single image det result.
Args:
det_result (list[list]): [[cls1_det, cls2_det, ...], ...].
The outer list indicates images, and the inner list indicates
per-class detected bboxes.
annotation (dict): Ground truth annotations where keys of
annotations are:
- bboxes: numpy array of shape (n, 4)
- labels: numpy array of shape (n, )
- bboxes_ignore (optional): numpy array of shape (k, 4)
- labels_ignore (optional): numpy array of shape (k, )
Returns:
float: mAP
"""
# use only bbox det result
if isinstance(det_result, tuple):
bbox_det_result = [det_result[0]]
else:
bbox_det_result = [det_result]
# mAP
iou_thrs = np.linspace(.5,
0.95,
int(np.round((0.95 - .5) / .05)) + 1,
endpoint=True)
mean_aps = []
for thr in iou_thrs:
mean_ap, _ = eval_map(bbox_det_result, [annotation],
iou_thr=thr,
logger='silent')
mean_aps.append(mean_ap)
return sum(mean_aps) / len(mean_aps)