Id |
Subject |
Object |
Predicate |
Lexical cue |
T385 |
0-210 |
Sentence |
denotes |
In total, 603 images for coughing, 634 images for hugging, 608 images for handshaking, and 623 images for door touching were used from COCO dataset [62] for transfer learning for the pre-trained model (YOLOv3). |
T386 |
211-289 |
Sentence |
denotes |
These images were taken from free sources found through Google image searches. |
T387 |
290-349 |
Sentence |
denotes |
For labelling objects, a semi-automatic method was applied. |
T388 |
350-393 |
Sentence |
denotes |
Darknet library was also used for training. |
T389 |
394-586 |
Sentence |
denotes |
For individual behaviors, all of the people in images were detected and labelled in a text file whilst the algorithm aggregated intersected bounding boxes of people into a single bounding box. |
T390 |
587-694 |
Sentence |
denotes |
As wrong labels might be generated, the images should be manually checked to correct misclassified objects. |
T391 |
695-788 |
Sentence |
denotes |
For this step 80 percent of the images were selected for training and 20 percent for testing. |
T392 |
789-867 |
Sentence |
denotes |
To increase the accuracy of this model, the configuration in Table 3 was used. |