Nowadays, the increased demand of a more affluent population urgently requires cost and time-effective phenotyping methodologies. The aim of this study is to generate a methodology based on Machine Learning techniques to high throughput phenotyping the biomass of soybean UAS-based 3D modeling. The approach will be developed in a novel and innovative way, ensuring ﬂexibility and simplicity in data acquisition, automation in the process and high-quality results, using low-cost sensors. The non-invasive methodology allows the determination of physiological growth dynamics by combining close-range photogrammetry and computer vision. During the growing season 2018, a soybean experiment was carried out at the Agronomy Center for Research and Education (ACRE) in West-Lafayette (Indiana, USA). Periodic images were acquired by G9X Canon camera on board senseFly eBee. The study area is reconstructed in 3D by Image-based modelling. Algorithms and techniques were combined to obtain the canopy volume to estimate the biomass per plot, training and validating the model by samples of AGB (Above Ground Biomass) and dry biomass tests. The accuracy of the results conﬁrm the possibility of biomass estimation in soybean by UAS-based SfM, affording a decision support tool for practical breeding.