Recognizing the Damaged Surface Parts of Cars in the Real Scene Using a Deep Learning Framework

Figure 4

A visual comparison between our model and some other techniques. (a) Original image. (b) Labeled image. (c) YOLOv3. (d) Mask RCNN. (e) VGG. (f) FFNN. (g) CNN. (h) Texture descriptor. (i) Proposed model.