Daryl Peralta

Daryl Peralta

University of the Philippines

Daryl Peralta is a researcher and lecturer at the Electrical and Electronics Engineering Institute of the University of the Philippines. He received his master’s degree from the same university where he worked on the problem of 3D reconstruction under the supervision of Prof. Rowel Atienza and Prof. Rhandley Cajote. During his MS, he developed the Scan-RL algorithm in the paper “Next-Best View Policy for 3D Reconstruction”. His research interests are computer vision, machine learning and robotics.

Next-Best View Policy for 3D Reconstruction

Manually selecting viewpoints or using commonly available flight planners like circular path for large-scale 3D reconstruction using drones often results in incomplete 3D models. Recent works have relied on hand-engineered heuristics such as information gain to select the Next-Best Views. In this work, we present a learning-based algorithm called Scan-RL to learn a Next-Best View (NBV) Policy. To train and evaluate the agent, we created Houses3K, a dataset of 3D house models. Our experiments show that using Scan-RL, the agent can scan houses with fewer number of steps and a shorter distance compared to our baseline circular path. Experimental results also demonstrate that a single NBV policy can be used to scan multiple houses including those that were not seen during training.