Empirical evidence has demonstrated that learning-based image compression can outperform classical compression frameworks. This has led to the ongoing standardization of learned-based image codecs, namely Joint Photographic Experts Group (JPEG) AI. The objective of JPEG AI is to enhance compression efficiency and provide a software and hardware-friendly solution. Based on our research, JPEG AI represents the first standardization that can facilitate the implementation of a learned image codec on a mobile device. This article presents an overview of the variable rate coding functionality in JPEG AI, which includes three variable rate adaptations: a three-dimensional quality map, a fast bit rate matching algorithm, and a training strategy. The variable rate adaptations offer a continuous rate function up to 2.0 bpp, exhibiting a high level of performance, a flexible bit allocation between different color components, and a region of interest function for the specified use case. The evaluation of performance encompasses both objective and subjective results. With regard to the objective bit rate matching, the main profile with low complexity yielded a 13.1% BD-rate gain over VVC intra, while the high profile with high complexity achieved a 19.2% BD-rate gain over VVC intra. The BD-rate result is calculated as the mean of the seven perceptual metrics defined in the JPEG AI common test conditions. With respect to subjective results, the example of improving the quality of the region of interest is illustrated.

Virtual-to-real registration is a crucial aspect of 3D registration, which presents a more challenging multimodal 3D registration problem due to the different data structures between virtual and real models. In this paper, we utilize point cloud registration algorithm to align virtual and real models, transforming the multimodal 3D registration problem into a cross-source point cloud registration problem. We propose a method for extracting macro and micro structures to represent the shared features of virtual and real models, combined with a multi-constraint registration algorithm, to achieve high-accuracy virtual-to-real registration tasks. This method can register unseen 3D objects using virtual prior knowledge, and allow partial point cloud registration without the need for a 360-degree scan of the model. Our approach can effectively resist interference from typical cross-source point cloud registration problems such as varying densities, missing data, and distribution changes. Furthermore, by processing only 0.2% of the original number of point clouds through downsampling, we can effectively diminish the effects of noise and outlier, as well as significantly decrease time consuming. Experimental results show that our algorithm outperforms other advanced point cloud registration algorithms in cross-source point cloud registration for virtual-to-real registration.