Traditional urban design is time-consuming and laborious. We propose a computer-generated architecture (CGA)-based workflow in this work, with the goal of allowing designers to take advantage of a high level of automation. This workflow is based on procedural modeling. A three-step CGA rule was applied to implement 3D urban procedural modeling, (1) parcel subdivision and clustering, (2) building extrusion, and (3) texture mapping. Parcel subdivision and clustering is the key step of layout modeling, giving the modeler flexibility to adjust the placement and size of the inner building lots. Subsequently, a land-use-based combination of eight common building types and layouts was used to generate various urban forms for different urban functional zones. Finally, individual buildings were decorated by creating texture maps of a planar section of the building facade or, alternatively, decomposing facades into sets of repeating elements and texture maps. We employed the proposed workflow in the H-village urban redevelopment program and an air–rail integration zone development program in Guangzhou. Three design proposals were generated for each project. The results demonstrated that this workflow could generate multiple layout proposals and alternative facade textures quickly and, therefore, address most of the collaborative issues with its analysis functions, including a flexible adjustment mechanism and real-time visualization.

Point cloud reconstruction is an ingredient in geometry modeling, computer graphics, and 3D vision. In this paper, we propose a novel unsupervised learning method called the Recurrent Multi-Step Moving Strategy, which progressively moves query points toward the underlying surface to accurately learn unsigned distance fields (UDFs) for point cloud reconstruction. Specifically, we design a recurrent network for UDF estimation that integrates a multi-step strategy for query movement. This model treats query movement as a trajectory prediction process, establishing dependencies between the current query move decision and the previous path, thus utilizing temporal information to improve UDF estimation accuracy. Further, we design distance and gradient regularization losses to ensure the precision, consistency, and continuity of the estimated UDFs. Extensive evaluations, comparisons, and ablation studies are conducted to show the superiority of our method over the competing approaches in terms of reconstruction accuracy and generality. Our unsupervised reconstruction method outperforms many supervised techniques and demonstrates efficacy across diverse scenarios, including single-object, indoor, and outdoor benchmarks.

There are a prohibitively large number of floating-point time series data generated at an unprecedentedly high rate. An efficient, compact and lossless compression for time series data is of great importance for a wide range of scenarios. Most existing lossless floating-point compression methods are based on the XOR operation, but they do not fully exploit the trailing zeros, which usually results in an unsatisfactory compression ratio. This paper proposes an Erasing-based Lossless Floating-point compression algorithm, i.e., Elf. The main idea of Elf is to erase the last few bits (i.e., set them to zero) of floating-point values, so the XORed values are supposed to contain many trailing zeros. The challenges of the erasing-based method are three-fold. First, how to quickly determine the erased bits? Second, how to losslessly recover the original data from the erased ones? Third, how to compactly encode the erased data? Through rigorous mathematical analysis, Elf can directly determine the erased bits and restore the original values without losing any precision. To further improve the compression ratio, we propose a novel encoding strategy for the XORed values with many trailing zeros. Furthermore, observing the values in a time series usually have similar significand counts, we propose an upgraded version of Elf named Elf+ by optimizing the significand count encoding strategy, which improves the compression ratio and reduces the running time further. Both Elf and Elf+ work in a streaming fashion. They take only O(N) (where N is the length of a time series) in time and O(1) in space, and achieve a notable compression ratio with a theoretical guarantee. Extensive experiments using 22 datasets show the powerful performance of Elf and Elf+ compared with 9 advanced competitors for both double-precision and single-precision floating-point values.

MVS (Multi-View Stereo) establishes dense correspondences among multiple calibrated images to generate 3D point clouds, which has broad applications in fields such as 3D modeling, robot navigation, and autonomous driving. In street view MVS, the distant, weakly-textured sky pixels in images significantly degrade the quality of the generated point cloud, manifesting as pronounced edge noise at building boundaries, and the completeness of the point cloud requires further improvement. Therefore, we design an Edge-enhanced Sky Masking Module to free street view MVS from sky interference, reducing edge noise by approximately 40%. In addition, we propose a Fusion Module based on Local Planarity Features, which integrates the strengths of both traditional and learning-based algorithms to generate superior dense point clouds, outperforming current mainstream methods in terms of completeness and F1 score.