This paper presents CrePoster, a data-driven framework to generate aesthetic posters for Chinese cultural relics, aiming to enhance the exhibition experience and promote cultural spread. CrePoster comprises three modules: (1) object segmentation module, (2) content generation module, and (3) poster generation module. Upon processing a cultural relic image, the object segmentation module first leverages a cascaded U2Net-SAM structure to obtain the visual target. Secondly, the content generation module utilizes a multi-target learning-enabled caption generator to produce professional captions. Thirdly, the Multimodal Large Language Model (MLLM) based poster generation module adaptively creates aesthetic parameters, including layout and color scheme, ultimately rendering them into refined posters.

Path planning is the core technology of mobile robot decision‐making and control and is also a research hotspot in the field of artificial intelligence. Aiming at the problems of slow response speed, long planning path, unsafe factors, and a large number of turns in the conventional path planning algorithm, an improved multiobjective genetic algorithm (IMGA) is proposed to solve static global path planning in this paper. The algorithm uses a heuristic median insertion method to establish the initial population, which improves the feasibility of the initial path and generates a multiobjective fitness function based on three indicators: path length, path security, and path energy consumption, to ensure the quality of the planned path. Then, the selection, crossover, and mutation operators are designed by using the layered method, the single‐point crossover method, and the eight‐neighborhood‐domain single‐point mutation method, respectively. Finally, the delete operation is added, to further ensure the efficient operation of the mobile robot. Simulation experiments in the grid environment show that the algorithm can improve the defects of the traditional genetic algorithm (GA), such as slow convergence speed and easy to fall into local optimum. Compared with GA, the optimal path length obtained by planning is shortened by 17%.

3D High Efficiency Video Coding (3D-HEVC) is an extension of the HEVC standard for coding of multiview videos and depth maps. It inherits the same quadtree coding structure as HEVC for both components, which allows recursively splitting into four equal-sized coding units (CU). One of 11 different prediction modes is chosen to code a CU in inter-frames. Similar to the joint model of H.264/AVC, the mode decision process in HM (reference software of HEVC) is performed using all the possible depth levels and prediction modes to find the one with the least rate distortion cost using a Lagrange multiplier. Furthermore, both motion estimation and disparity estimation need to be performed in the encoding process of 3D-HEVC. Those tools achieve high coding efficiency, but lead to a significant computational complexity. In this article, we propose a fast mode decision algorithm for 3D-HEVC. Since multiview videos and their associated depth maps represent the same scene, at the same time instant, their prediction modes are closely linked. Furthermore, the prediction information of a CU at the depth level X is strongly related to that of its parent CU at the depth level X-1 in the quadtree coding structure of HEVC since two corresponding CUs from two neighboring depth levels share similar video characteristics. The proposed algorithm jointly exploits the inter-view coding mode correlation, the inter-component (texture-depth) correlation and the inter-level correlation in the quadtree structure of 3D-HEVC. Experimental results show that our algorithm saves 66% encoder runtime on average with only a 0.2% BD-Rate increase on coded views and 1.3% BD-Rate increase on synthesized views.