Foot Depth Map Point Cloud Completion using Deep Learning with Residual Blocks

Fit is extremely important in footwear as fit largely determines performance and comfort. Current footwear fit estimation mainly uses only shoe size, which is extremely limited in characterizing the shape of a foot or the shape of a shoe. 3D scanning presents a solution to this, where a foot shape can be captured and virtually fit with shoe models. Traditional 3D scanning techniques have their own complications however, stemming from their need to collect views covering all aspects of an object. In this work we explore a deep learning technique to compete a foot scan point cloud from information contained in a single depth map view. We examine the benefits of implementing residual blocks in architectures for this application, and find that they can improve accuracies while reducing model size and training time.