The field of 3D reconstruction and point cloud processing is rapidly advancing, with a focus on developing more efficient and accurate methods for processing large-scale datasets. One of the key trends in this area is the use of transformer-based architectures, which have shown significant improvements in performance and robustness. These models are being applied to a range of tasks, including 3D reconstruction from multi-view images, point cloud completion, and rotation estimation. Another important area of research is the development of methods for handling missing or noisy data, such as outlier view rejection and motion-aware updating. Notable papers in this area include HeartFormer, which proposes a novel point cloud completion network for 3D four-chamber cardiac reconstruction, and FlashVGGT, which introduces an efficient alternative to traditional visual geometry grounding transformers. Overall, these advances are enabling more accurate and efficient 3D reconstruction and point cloud processing, with potential applications in a range of fields, including computer vision, robotics, and healthcare. Noteworthy papers: HeartFormer proposes a novel point cloud completion network for 3D four-chamber cardiac reconstruction. FlashVGGT introduces an efficient alternative to traditional visual geometry grounding transformers, achieving competitive reconstruction accuracy while reducing inference time.