The field of video and image processing is witnessing significant advancements, driven by the development of novel deep learning-based methods and the integration of traditional techniques with modern approaches. A key direction in this field is the improvement of video inpainting, denoising, and compression technologies, with a focus on achieving high-quality outcomes, ensuring temporal consistency, and managing complex object interactions. Another area of innovation is the removal of shadows from images and videos, where new methods leveraging frequency characteristics and wavelet-based image decomposition are showing promising results. Furthermore, the creation of large-scale datasets for specific applications, such as geometry-aware shadow detection in remote sensing, is facilitating the training and evaluation of more accurate models. Noteworthy papers in this area include VIP, which introduces a novel promptless video inpainting framework, and FASR-Net, which proposes an unsupervised frequency-aware shadow removal network. Additionally, papers like 3DM-WeConvene and FANeRV are making significant contributions to learned image compression and neural representations for video, respectively.