Previously, machine learning and computer vision (remote sensing) have been used to classify crops in agriculture from imagery. Additionally, a recent paper at ECCV presents a novel method of obtaining 3D shape, viewpoint, and texture of objects in imagery without keypoints/annotations, viewpoint ground truths, or shape hints needed. In this ongoing work, we propose a pipeline that first classifies  crops in images and then forms predicted 3D models of them without supervision. This novel approach provides farmers with enhanced information about the food they grow and addresses the issue of hunger in developing countries in a computational manner.