Research/ SPACE AI & AUTONOMY
DreamSat-2.0: Towards a General Single-View Asteroid 3D Reconstruction
Santiago Diaz, Xinghui Hu, Josiane Uwumukiza, Giovanni Lavezzi, Victor Rodriguez-Fernandez, Richard Linares
Perception & State EstimationarXiv: cs.CV · Learning-based MethodsarXiv: cs.LG · Simulation, Digital Twins & Benchmarkstext match
arXivcs.CV · cs.LGOpenAlex3D Shape Modeling and Analysis · Advanced Vision and Imaging · Generative Adversarial Networks and Image Synthesis
To enhance asteroid exploration and autonomous spacecraft navigation, we introduce DreamSat-2.0, a pipeline that benchmarks three state-of-the-art 3D reconstruction models-Hunyuan-3D, Trellis-3D, and Ouroboros-3D-on custom spacecraft and asteroid datasets. Our systematic analysis, using 2D perceptual (image quality) and 3D geometric (shape accuracy) metrics, reveals that model performance is domain-dependent. While models produce higher-quality images of complex spacecraft, they achieve better geometric reconstructions for the simpler forms of asteroids. New benchmarks are established, with Hunyuan-3D achieving top perceptual scores on spacecraft but its best geometric accuracy on asteroids, marking a significant advance over our prior work.