4 papers - avg viability 5.3
A real-time lunar surface mapping framework using 3D Gaussian Splatting and dense perception models for centimeter-level geometric accuracy.
An efficient RGB-D scene understanding model that performs semantic segmentation, instance segmentation, orientation estimation, panoptic segmentation, and scene classification, outperforming existing methods in accuracy and speed.
A deep learning method for robust visual feature tracking in challenging environments, integrated into a VIO system.
Develop a monocular neural tactile field reconstruction tool for improved robotic navigation through deformable environments.