small updates to scripts

Author: ngr-francesco

.gitignore

@@ -0,0 +1,3 @@
+data/dataset/
+checkpoints/
+__pycache__

core/utils.py

@@ -143,22 +143,23 @@ def draw_bbox(image, bboxes, classes=read_class_names(cfg.YOLO.CLASSES), show_la
         coor[2] = int(coor[2] * image_h)
         coor[1] = int(coor[1] * image_w)
         coor[3] = int(coor[3] * image_w)
+        coor = coor.astype(int)
 
         fontScale = 0.5
         score = out_scores[0][i]
         class_ind = int(out_classes[0][i])
         bbox_color = colors[class_ind]
         bbox_thick = int(0.6 * (image_h + image_w) / 600)
         c1, c2 = (coor[1], coor[0]), (coor[3], coor[2])
-        cv2.rectangle(image, c1, c2, bbox_color, bbox_thick)
+        cv2.rectangle(img = image, pt1 = c1, pt2= c2, color=bbox_color, thickness= bbox_thick)
 
         if show_label:
             bbox_mess = '%s: %.2f' % (classes[class_ind], score)
             t_size = cv2.getTextSize(bbox_mess, 0, fontScale, thickness=bbox_thick // 2)[0]
-            c3 = (c1[0] + t_size[0], c1[1] - t_size[1] - 3)
-            cv2.rectangle(image, c1, (np.float32(c3[0]), np.float32(c3[1])), bbox_color, -1) #filled
+            c3 = (c1[0] + t_size[0], c1[1] + t_size[1] + 5)
+            cv2.rectangle(img= image, pt1= c1, pt2=(int(c3[0]), int(c3[1])), color=bbox_color, thickness= -1) #filled
 
-            cv2.putText(image, bbox_mess, (c1[0], np.float32(c1[1] - 2)), cv2.FONT_HERSHEY_SIMPLEX,
+            cv2.putText(image, bbox_mess, (int(c1[0]), int(c1[1] + t_size[1] + 2)), cv2.FONT_HERSHEY_SIMPLEX,
                         fontScale, (0, 0, 0), bbox_thick // 2, lineType=cv2.LINE_AA)
     return image
 

detect.py

@@ -50,20 +50,29 @@ def main(_argv):
         interpreter.allocate_tensors()
         input_details = interpreter.get_input_details()
         output_details = interpreter.get_output_details()
-        print(input_details)
-        print(output_details)
+        for key,value in input_details[0].items():
+            print(key,value)
+        for key,value in output_details[0].items():
+            print(key,value)
         interpreter.set_tensor(input_details[0]['index'], images_data)
         interpreter.invoke()
         pred = [interpreter.get_tensor(output_details[i]['index']) for i in range(len(output_details))]
+        print("PREDICTION", np.shape(pred[0]))
         if FLAGS.model == 'yolov3' and FLAGS.tiny == True:
             boxes, pred_conf = filter_boxes(pred[1], pred[0], score_threshold=0.25, input_shape=tf.constant([input_size, input_size]))
         else:
-            boxes, pred_conf = filter_boxes(pred[0], pred[1], score_threshold=0.25, input_shape=tf.constant([input_size, input_size]))
+            pred_bbox = pred
+            for value in pred_bbox:
+                boxes = value[:, :, 0:4]
+                pred_conf = value[:, :, 4:]
+            # boxes, pred_conf = filter_boxes(pred[0], pred[1], score_threshold=0.25, input_shape=tf.constant([input_size, input_size]))
     else:
         saved_model_loaded = tf.saved_model.load(FLAGS.weights, tags=[tag_constants.SERVING])
         infer = saved_model_loaded.signatures['serving_default']
         batch_data = tf.constant(images_data)
         pred_bbox = infer(batch_data)
+        for key,value in pred_bbox.items():
+            print(f"{key}: {np.shape(value)}")
         for key, value in pred_bbox.items():
             boxes = value[:, :, 0:4]
             pred_conf = value[:, :, 4:]

result.png

@@ -7,15 +7,14 @@
 import cv2
 
 flags.DEFINE_string('coco_data', './val2017.pkl', 'path to coco data')
-flags.DEFINE_string('classes', '../data/classes/coco.names', 'path to classes file')
+flags.DEFINE_string('classes', '../classes/coco.names', 'path to classes file')
 flags.DEFINE_string('coco_path', "/Volumes/Elements/data/coco_dataset/coco", 'resize images to')
 flags.DEFINE_string('image_path', "images/val2017", 'path to image val')
-flags.DEFINE_string('anno_path_val', '../data/dataset/val2017.txt', 'path to classes file')
+flags.DEFINE_string('anno_path_val', './val2017.txt', 'output file')
 
 def convert_annotation(output, data, data_type = "val"):
     class_names = [c.strip() for c in open(FLAGS.classes).readlines()]
     replace_dict = {"couch": "sofa", "airplane": "aeroplane", "tv": "tvmonitor", "motorcycle": "motorbike"}
-
     if os.path.exists(output): os.remove(output)
     directory_path = os.path.join(FLAGS.coco_path, FLAGS.image_path)
     # if data_type == "train":