An Evaluation of Deep Learning Methods for Small Object Detection

Figure 5

Highlight of bounding boxes from comparative backbones on small object dataset. We here select YOLO with Darknet-53 and ResNet-50 for objective comparison because there have obviously the same layers in their networks along with the significant techniques such as skip connections and residual blocks. The bounding boxes show that ResNet-50 has the sensitivity to areas which resembles the objects of interest than Darknet-53. Similarly, ResNet-50-FPN and ResNet-50-C4 are chosen to consider. The detection shows that combining ResNet-50 with FPN outputs a better performance rather than the original one. Particularly, misdetection happens in more density than ResNet-50-FPN such as in columns 4 and 5. Zoom in to see more detail.

(a)

(b)

(c)

(d)