The field of egocentric vision and 3D scene reconstruction is rapidly advancing, with a focus on developing more accurate and robust models for understanding and interacting with complex environments. Recent research has emphasized the importance of multimodal and multi-perspective approaches, incorporating data from various sources such as wearable devices, cameras, and sensors. This has led to significant improvements in tasks like action recognition, human-centric perception, and 6-DoF navigation. Noteworthy papers in this area include EgoExOR, which introduces a comprehensive dataset for surgical activity understanding, and EyeNavGS, which provides a large-scale dataset for 6-DoF navigation in virtual reality. Additionally, the Oxford Day-and-Night dataset offers a unique platform for benchmarking egocentric 3D vision under challenging lighting conditions.