An evaluation of VGG16 and YOLO v3 on hand-drawn images
Lee Hoang · 2019
This thesis evaluates the accuracy and performance of VGG16, a convolutional neural network (CNN), and YOLO v3, an object detector, on a dataset of 1000 hand-drawn images. Unlike with photographs, which possess high amounts of detail, sketches tend to lack much detail aside from the freehand lines that comprise them. This is further detailed in prior works about Sketch-based Image Retrieval (SBIR), a classification task to map photographs to sketches; and SketchParse, a CNN that analyzes sketch features and assigns captions. In this paper, I show the differences in classification accuracy between VGG16 and YOLO v3. The former model, pretrained on ImageNet, showed a test accuracy as high as 79.6%. On the other hand, YOLO v3, pretrained on MS COCO, performed worse; it misclassified objects in the dog category and across all categories, made no detections on several images.