Enhancing Object Detection Robustness for Cross-Depiction Through Neural Style Transfer

Modern neural networks models for computer vision are trained on millions of images. The idea is that models are able to increase generalization when the dataset contains well diversified images, e.g. with varied illumination and environmental conditions of the same objects. Generalization is particularly relevant in object detection, especially for what concerns the cross-depiction problem. In this work we explore the use of Neural Style Transfer as a novel technique to morph the original data, with the aim to enhance model generalization. To verify the effect on performances for object detection models, we selected the Faster R-CNN model to be applied on the Pascal VOC 2012 dataset. A number of tests were performed through style variations on images and by tuning Neural Style Transfer parameters to maintain the content of the original images. The experiments showed promising results, which effectively provide a foundation for future studies on cross-depiction via Neural Style Transfer.