Aiming at the lack of high-speed and high-precision planar line segment distance solving algorithms in the existing CAD geometry engine and VLSI design, we propose a planar line segment distance solving algorithm based on local coordinate system. First, a local coordinate system is established by coordinate transformation; then, according to the relative position relationship between two line segments, the planar line segment distance solving is classified into 12 kinds of nearest point solving cases; finally, the planar line segment distance is solved by computing the distance of the nearest points. The experiments are carried out with PCB line segments data and randomly generated line segments data, and the experiment results show that the algorithm is 23% to 35% more efficient than the compared algorithms. Moreover, the algorithm can still control the computational error in four critical cases, which proves its effectiveness.
1 December 2023 in Journal of Computer-Aided Design & Computer Graphics
Current CAD software is slow to load large STEP neutral files, and its efficiency bottleneck lies in its analytic algorithm. The parsing algorithm based on single thread cannot meet the requirements of CAD application. This paper proposes a hierarchical parallel data parsing algorithm based on multithreading, which makes use of the multi-core parallel architecture adopted by current mainstream CPUs, and greatly improves the efficiency of STEP file parsing. According to the hierarchical characteristics of data in the model, the parallel acceleration of data parsing at each layer is carried out by using thread pool. The algorithm in this paper has been implemented in open source geometry engine OpenGE. In this paper, comparison between our algorithm and OpenCASCADE is performed to verify the running effects of the parallel algorithms. The experimental results show that the running speed of the parallel parsing program is about one times faster than that of the serial parsing program based on single thread and about one and a half times faster than that of OpenCASCADE.
To address the prominent challenges of insufficient global attribute preservation and text-image semantic conflicts in virtual dressing tasks, this paper proposes a virtual dressing framework driven by dual-stream encoding and structured semantic disentanglement. Specifically, a dual-stream garment encoder is first introduced to extract holistic semantic information alongside local texture details, thereby constructing semantically-guided, multi-scale garment features. On this basis, a structured semantic disentanglement module integrates a zero-shot classifier with a Large Language Model (LLM) to perform category-guided de-attribution and semantic aggregation on the textual conditions, effectively mitigating the interference of text-image semantic mismatches during the generative process. Finally, a cascading attention mechanism is employed to achieve an orderly fusion of human structural priors, garment features, and textual prompts. Extensive experiments demonstrate that the proposed method delivers state-of-the-art comprehensive performance across multiple benchmark datasets, significantly enhancing the preservation of global garment attributes while successfully suppressing the negative impacts of cross-modal semantic conflicts.