The field of computer vision and machine learning is rapidly advancing, with a focus on developing innovative solutions for real-world applications. Recent research has centered on improving the accuracy and efficiency of depth estimation, object detection, and segmentation, with a particular emphasis on enabling these technologies to operate in real-time. This has significant implications for fields such as robotics, autonomous driving, and assistive technologies for the visually impaired. Notably, the development of semi-supervised and unsupervised learning frameworks has allowed for the creation of more robust and adaptable models, capable of generalizing across diverse datasets and environments. Furthermore, the integration of transfer learning and quantization techniques has facilitated the deployment of these models on embedded systems, enabling their use in a wide range of applications. Overall, the current direction of the field is towards the development of more efficient, accurate, and scalable computer vision and machine learning solutions that can be applied in real-world contexts. Notable papers in this area include: SSSUMO, which introduces a semi-supervised deep learning approach for submovement decomposition that achieves state-of-the-art accuracy and speed. Prompt2DEM, which presents a framework for the estimation of high-resolution Digital Elevation Models (DEMs) using a monocular foundation model, achieving a 100x resolution gain and surpassing prior methods by an order of magnitude.