This paper presents a workflow for 3D reconstruction of archaeological objects and contexts using a low-cost UAV for data capture and Structure from Motion (SfM) algorithms for 3D reconstruction. The goal is to evaluate the feasibility of using affordable hardware instead of expensive equipment. The workflow consists of three main phases: generating binary masks to select regions of interest, producing dense point clouds, and transforming them into a triangular mesh. The use of binary masks is critical for improving the visual quality of the generated model. The paper proposes a semi-automated method for creating binary masks and a low-cost method for generating triangular meshes of archaeological objects and contexts. The study uses three pre-Hispanic monoliths as case studies. The results show that using binary masks significantly reduces noise in the point cloud and improves the quality of the triangular mesh. The paper also discusses the challenges of evaluating the quality of virtual reconstructions, as archaeological objects often lack regular surfaces or have not been previously reconstructed. The study highlights the importance of building robust datasets with a large number of images and both full object and detailed relief images to increase the detail level of the point cloud and triangular mesh. The paper concludes that a robust methodology for 3D reconstruction of archaeological contexts can be developed using low-cost devices and open-source software. The results demonstrate that the proposed method is effective in preserving details such as reliefs and colors from the original object. The study also emphasizes the need for quantitative evaluation frameworks to assess the reliability of generated models.This paper presents a workflow for 3D reconstruction of archaeological objects and contexts using a low-cost UAV for data capture and Structure from Motion (SfM) algorithms for 3D reconstruction. The goal is to evaluate the feasibility of using affordable hardware instead of expensive equipment. The workflow consists of three main phases: generating binary masks to select regions of interest, producing dense point clouds, and transforming them into a triangular mesh. The use of binary masks is critical for improving the visual quality of the generated model. The paper proposes a semi-automated method for creating binary masks and a low-cost method for generating triangular meshes of archaeological objects and contexts. The study uses three pre-Hispanic monoliths as case studies. The results show that using binary masks significantly reduces noise in the point cloud and improves the quality of the triangular mesh. The paper also discusses the challenges of evaluating the quality of virtual reconstructions, as archaeological objects often lack regular surfaces or have not been previously reconstructed. The study highlights the importance of building robust datasets with a large number of images and both full object and detailed relief images to increase the detail level of the point cloud and triangular mesh. The paper concludes that a robust methodology for 3D reconstruction of archaeological contexts can be developed using low-cost devices and open-source software. The results demonstrate that the proposed method is effective in preserving details such as reliefs and colors from the original object. The study also emphasizes the need for quantitative evaluation frameworks to assess the reliability of generated models.