The current research on text-guided 3D synthesis predominantly utilizes complex diffusion models, posing significant challenges in tasks like terrain generation. This study ventures into the direct synthesis of text-to-3D terrain in a zero-shot fashion, circumventing the need for diffusion models. By exploiting the large language model’s inherent spatial awareness, we innovatively formulate a method to update existing 3D models through text, thereby enhancing their accuracy. Specifically, we introduce a Gaussian–Voronoi map data structure that converts simplistic map summaries into detailed terrain heightmaps. Employing a chain-of-thought behavior tree approach, which combines action chains and thought trees, the model is guided to analyze a variety of textual inputs and extract relevant terrain data, effectively bridging the gap between textual descriptions and 3D models. Furthermore, we develop a text–terrain re-editing technique utilizing multiagent reasoning, allowing for the dynamic update of the terrain’s representational structure. Our experimental results indicate that this method proficiently interprets the spatial information embedded in the text and generates controllable 3D terrains with superior visual quality.

This paper introduces a variational autoencoder network designed for video colorization using reference images, addressing the challenge of colorizing black-and-white videos. Although recent techniques perform well in some scenarios, they often struggle with color inconsistencies and artifacts in videos that feature complex scenes and long durations. To tackle this, we propose a variational autoencoder framework that incorporates spatio-temporal information for efficient video colorization. To improve temporal consistency, we unify semantic correspondence with color propagation, allowing for simultaneous guidance in colorizing grayscale video frames. Additionally, the variational autoencoder learns spatio-temporal feature representations by mapping video frames into a latent space through an encoder network. The decoder network then transforms these latent features back into color images. Compared to traditional coloring methods, our approach accurately captures temporal relationships between video frames, providing precise colorization while ensuring video consistency. To further enhance video quality, we apply a specialized loss function that constrains the generated output, ensuring that the colorized video remains spatio-temporally consistent and natural. Experimental results demonstrate that our method significantly improves the video colorization process.

The emerging fast reconstruction method 3D Gaussian Splatting, based on 3D Gaussian and point cloud, utilizes an explicit representation approach, demonstrating excellent image quality and outstanding real-time performance, effectively addressing the challenges of implementing Neural Radiance Field (NeRF) inference on mobile devices. To achieve real-time rendering of 3D reconstruction on mobile devices, the 3D Gaussian Splatting Radiance Field model has been improved and optimized to save computational resources while maintaining rendering quality. Firstly, computational cost is reduced by employing Dual Splatting, thereby alleviating the burden of high memory consumption. Secondly, by reducing the processing of low-contribution points, the generation of redundant data is minimized. Additionally, thread saturation overflow operations are introduced to accelerate the overall convergence of the original Tile-based Rasterizer method. Finally, voxelized point clouds are used to further reduce the rendering and training overhead, effectively addressing the limited hardware resources on mobile devices. Experimental results on multiple common datasets indicate that the improved method reduces the average maximum memory occupation during training by 33.30% and reduces the average space occupancy of the model by 49.35%. For large-scale scene datasets, the improved method achieves an average reduction of 37.08% in model space occupancy while ensuring the quality of rendering.