The field of 3D reconstruction and robot exploration is rapidly advancing, with a focus on improving the accuracy and efficiency of these tasks. Recent research has led to the development of new methods for 3D reconstruction, such as the use of implicit neural fields and hierarchical uncertainty quantification, which have shown significant improvements in terms of accuracy and completeness. Additionally, there has been a growing interest in active exploration strategies, which enable robots to selectively gather more information about their environment and improve their understanding of the scene. These advancements have the potential to enable more effective and efficient robot exploration and 3D reconstruction in a variety of applications, including search and rescue, environmental monitoring, and robotic manipulation. Noteworthy papers include:

• QAL: A Loss for Recall Precision Balance in 3D Reconstruction, which proposes a new loss function for 3D reconstruction that balances recall and precision.
• Active3D: Active High-Fidelity 3D Reconstruction via Hierarchical Uncertainty Quantification, which presents an active exploration framework for high-fidelity 3D reconstruction that selects next-best-views through an uncertainty-driven motion planner.
• Quality-guided UAV Surface Exploration for 3D Reconstruction, which proposes a novel modular Next-Best-View planning framework for aerial robots that explicitly uses a reconstruction quality objective to guide the exploration planning.