inference_example.py

import keras

# import keras_retinanet
from object_detector_retinanet.keras_retinanet import models
from object_detector_retinanet.keras_retinanet.utils.image import read_image_bgr, preprocess_image, resize_image
from object_detector_retinanet.keras_retinanet.utils.visualization import draw_box, draw_caption
from object_detector_retinanet.keras_retinanet.utils.colors import label_color

# import for EM Merger and viz
from object_detector_retinanet.keras_retinanet.utils import EmMerger
from object_detector_retinanet.utils import create_folder, root_dir


# import miscellaneous modules
import matplotlib.pyplot as plt
import cv2
import os
import numpy as np
import time

# load label to names mapping for visualization purposes
labels_to_names = {0: 'object'}

model_path = # PATH TO MODEL
model = models.load_model(model_path, backbone_name='resnet50', convert=True)
image_path = #PATH TO TEST IMAGE
image = read_image_bgr(image_path)

# for filtering predictions based on score (objectness/confidence)
threshold = 0.3

# copy to draw on
draw = image.copy()
draw = cv2.cvtColor(draw, cv2.COLOR_BGR2RGB)

# preprocess image for network
image = preprocess_image(image)
image, scale = resize_image(image)

# Run inference
boxes, hard_scores, labels, soft_scores = model.predict_on_batch(np.expand_dims(image, axis=0))
hard_score_rate = .3
max_detections = 9999

soft_scores = np.squeeze(soft_scores, axis=-1)
soft_scores = hard_score_rate * hard_scores + (1 - hard_score_rate) * soft_scores

# correct boxes for image scale
boxes /= scale

# select indices which have a score above the threshold
indices = np.where(hard_scores[0, :] > threshold)[0]

# select those scores
scores = soft_scores[0][indices]
hard_scores = hard_scores[0][indices]

# find the order with which to sort the scores
scores_sort = np.argsort(-scores)[:max_detections]

# select detections
image_boxes = boxes[0, indices[scores_sort], :]
image_scores = scores[scores_sort]
image_hard_scores = hard_scores[scores_sort]
image_labels = labels[0, indices[scores_sort]]
image_detections = np.concatenate(
    [image_boxes, np.expand_dims(image_scores, axis=1), np.expand_dims(image_labels, axis=1)], axis=1)
results = np.concatenate(
    [image_boxes, np.expand_dims(image_scores, axis=1), np.expand_dims(image_hard_scores, axis=1),
     np.expand_dims(image_labels, axis=1)], axis=1)
filtered_data = EmMerger.merge_detections(image_path, results)
filtered_boxes = []
filtered_scores = []
filtered_labels = []

csv_data_lst = []
csv_data_lst.append(['image_id', 'x1', 'y1', 'x2', 'y2', 'confidence', 'hard_score'])

for ind, detection in filtered_data.iterrows():
    box = np.asarray([detection['x1'], detection['y1'], detection['x2'], detection['y2']])
    filtered_boxes.append(box)
    filtered_scores.append(detection['confidence'])
    filtered_labels.append('{0:.2f}'.format(detection['hard_score']))
    row = [image_path, detection['x1'], detection['y1'], detection['x2'], detection['y2'],
           detection['confidence'], detection['hard_score']]
    csv_data_lst.append(row)

for box, score, label in zip(filtered_boxes, filtered_scores, filtered_labels):
    # scores are sorted so we can break
    if score < threshold:
        break

    color = [31, 0, 255]  # label_color(label) ## BUG HERE LABELS ARE FLOATS SO COLOR IS HARDCODED

    b = box.astype(int)
    draw_box(draw, b, color=color)

    caption = "{} {:.3f}".format(labels_to_names[0], score)
    draw_caption(draw, b, caption)

plt.figure(figsize=(20, 20))
plt.axis('off')
plt.imshow(draw)
plt.show()