Object detection device, monitoring device, training device, and model generation method

A training device comprising:

• an image data acquire to acquire image data indicating a training image;
• a first feature amount extractor to generate a first feature map using the image data;
• a second feature amount extractor to generate a second feature map using the image data and to generate a third feature map by performing addition or multiplication of the second feature map using the first feature map and weighting the second feature map;
• an object detector to detect an object in the training image using the third feature map;
• and
• a trainer to train the second feature amount extractor and the object detector in accordance with a detection result of the object detector, wherein
• a first feature amount in the first feature map uses a mid-level feature corresponding to objectness, and
• a second feature amount in the second feature map uses a high-level feature.

A model generation method comprising:

• acquiring image data indicating a training image;
• generating a first feature map using the image data;
• generating a second feature map using the image data and generating a third feature map by performing arithmetic operation of the second feature map using the first feature map and weighting the second feature map;
• detecting an object in the training image using the third feature map;
• and
• generating a machine learning model that receives the image data as input and outputs a detection result of the object by training a second feature amount extractor and an object detector in accordance with a detection result of the object detector, wherein
• a first feature amount in the first feature map uses a mid-level feature corresponding to objectness, and
• a second feature amount in the second feature map uses a high-level feature.