def display_differences(image,
gt_box,
gt_class_id,
gt_mask,
pred_box,
pred_class_id,
pred_score,
pred_mask,
class_names,
title="",
ax=None,
show_mask=True,
show_box=True,
iou_threshold=0.5,
score_threshold=0.5):
"""Display ground truth and prediction instances on the same image."""
# Match predictions to ground truth
gt_match, pred_match, overlaps = utils.compute_matches(
gt_box,
gt_class_id,
gt_mask,
pred_box,
pred_class_id,
pred_score,
pred_mask,
iou_threshold=iou_threshold,
score_threshold=score_threshold)
# Ground truth = green. Predictions = red
colors = [(0, 1, 0, .8)] * len(gt_match)\
+ [(1, 0, 0, 1)] * len(pred_match)
# Concatenate GT and predictions
class_ids = np.concatenate([gt_class_id, pred_class_id])
scores = np.concatenate([np.zeros([len(gt_match)]), pred_score])
boxes = np.concatenate([gt_box, pred_box])
masks = np.concatenate([gt_mask, pred_mask], axis=-1)
# Captions per instance show score/IoU
captions = ["" for m in gt_match] + [
"{:.2f} / {:.2f}".format(pred_score[i],
(overlaps[i, int(pred_match[i])] if
pred_match[i] > -1 else overlaps[i].max()))
for i in range(len(pred_match))
]
# Set title if not provided
title = title or "Ground Truth and Detections\n GT=green, pred=red, captions: score/IoU"
# Display
display_instances(image,
boxes,
masks,
class_ids,
class_names,
scores,
ax=ax,
show_bbox=show_box,
show_mask=show_mask,
colors=colors,
captions=captions,
title=title)
def compute_batch_metrics(model,
dataset,
inference_config,
image_ids,
verbose=False):
all_results = dict()
for image_id in image_ids:
print("Processing image id: %s" % image_id)
all_results[image_id] = dict()
image, image_meta, gt_class_ids, gt_boxes, gt_masks =\
modellib.load_image_gt(dataset, inference_config,
image_id=image_id, use_mini_mask=False)
all_results[image_id]['gt_class_ids'] = gt_class_ids
all_results[image_id]['gt_boxes'] = gt_boxes
all_results[image_id]['gt_masks'] = gt_masks
# Run object detection
r = model.detect([image], verbose=False)[0]
class_result = get_class_result(gt_class=gt_class_ids,
pred_class=r['class_ids'])
all_results[image_id]['class_result'] = class_result
all_results[image_id]['pred_class_ids'] = r['class_ids']
all_results[image_id]['pred_boxes'] = r['rois']
all_results[image_id]['pred_scores'] = r['scores']
all_results[image_id]['pred_masks'] = r['masks']
print("Ground Truth Classes: %s" % gt_class_ids)
print("Predicted Classes: %s" % r['class_ids'])
print("Predicted Scores: %s" % r['scores'])
print("Class Result: %s" % class_result)
if gt_masks.size > 0 and r['masks'].size > 0:
gt_match, pred_match, overlaps = modelutils.compute_matches(
gt_boxes=gt_boxes,
gt_class_ids=gt_class_ids,
gt_masks=gt_masks,
pred_boxes=r['rois'],
pred_class_ids=r['class_ids'],
pred_scores=r['scores'],
pred_masks=r['masks'],
iou_threshold=0.5,
score_threshold=0.0)
all_results[image_id]['gt_match'] = gt_match
all_results[image_id]['pred_match'] = pred_match
all_results[image_id]['overlaps'] = overlaps
elif gt_masks.size == 0 and r['masks'].size > 0:
all_results[image_id]['overlaps'] = np.zeros(
shape=(r['class_ids'].shape[0], 1))
else:
all_results[image_id]['overlaps'] = np.array([])
print("Overlap: \n%s" % all_results[image_id]['overlaps'])
print("\n")
return all_results