CN 11-5366/S     ISSN 1673-1530
“风景园林,不只是一本期刊。”

基于网络大数据与AIGC微调模型的风景园林辅助设计工作流探索——以绿道为例

Landscape Architecture Aided Design Workflow Based on Big Data and AIGC Fine-tuning Model: A Case Study of Greenway

  • 摘要: 【目的】我国城镇化建设进入质量提升阶段,风景园林设计实践呈现出细分化、定制化、专业化的发展趋势,在大模型基础上如何结合特定需求开展微调训练,针对细分场景辅助设计师开展高效设计决策具有重要价值。【方法】采用设计进行研究(Research for Design, RfD)方法,基于网络大数据与AIGC微调模型技术,构建“大数据收集-智能评价-微调模型构建-设计辅助生成”的风景园林辅助设计工作流程框架,并对绿道场景进行实例应用。【结果】通过应用工作流,成功获得针对绿道场景的生成微调模型,并提出“强控制”与“弱控制”两大场景生成控制的有效范式。【结论】在此基础上,探讨工作流的适应潜力与局限性,为未来风景园林人工智能辅助设计高效研究决策提供参考。

     

    Abstract: Objective As China''s urbanization construction enters the stage of quality improvement, landscape architecture design practice presents the development trend of segmentation, customization and specialization. Designers are needed to study the design experience in segmented scenarios and come up with customized design strategies. The rapid development of the internet industry and artificial intelligence technology is transforming the traditional working modes of landscape architecture design, demonstrating immense potential in the fields of design evaluation research and computer-aided design. Against this backdrop, the research aims to explore the path of AI-aided design for customized landscape architecture scenarios. By combining network big data and fine-tuning model technology of AI Generated Content (AIGC), the research aims to bridge the path between big data analysis research and scenario generation in design practice, and construct a lightweight Landscape Architecture AI-aided design workflow to address the trends of segmentation, specialization, and customization in landscape architecture design practice. Methods The research adopts the Research for Design (RfD) methodology to build a workflow framework for integrating research and design practice. The workflow can be divided into four major processes: network big data collection, intelligent evaluation of network big data, AIGC image fine-tuning model construction, and AI-aided design generation. 1) Network big data collection: obtain data sets related to the required landscape architecture segmentation scenarios through online social platforms for evaluation and fine-tuning model training. 2) Intelligent evaluation of network big data: analyze and categorize image data, and filter out the scene images with excellent user evaluation by evaluating the text sentiment evaluation and subsidiary information analysis. 3) AIGC image fine-tuning model construction: utilize the high-quality image dataset obtained in the previous stage to conduct fine-tuning model training based on a mature pre-trained general model. Inject relevant knowledge and experience from the sub-scenes of landscape architecture in a cost-effective manner, thereby enhancing the model''s generative capabilities. 4) AI-aided design generation: employ the fine-tuned model obtained through training to assist in generating scene images according to the needs of design practice. Based on the intensity of control over the generated content, divide the scene-aided generation into "weakly controlled" and "strongly controlled" aided design scenarios. The research takes greenways as a typical case to verify the performance of the workflow framework. Image-text data related to Beijing greenways from 2013-2022 were collected from Weibo platform as the original dataset. Image features are extracted using the pre-trained convolutional neural network model InceptionResNetV2 and the image data is clustered by K-Means clustering algorithm. Through the process of image recognition and clustering, a total of 11 categories of images were obtained, including signage systems, pathways (cycling lanes, pedestrian walkways, waterside promenades), recreational facilities (children''s playgrounds, elderly activity areas, fitness areas, rest benches), flora (flowers), forested hills, trees, water bodies (hardscape and softscape waterside spaces), sculptures, plazas, and pavilions/towers/halls. Among these, the image categories of pathways, water bodies, and plazas were selected as they are representative of greenway spatial scenarios. Consequently, these three groups were chosen as data samples for subsequent analysis and model training. Using SnowNLP to analyze the sentiment of the text associated with the image, obtain the sentiment score, and correct the score by combining the microblog interaction data. Based on the corrected sentiment score to judge the image quality, the image data is gradually cleaned and filtered into a training set for LoRA fine-tuning model training. Results The LoRA model for the greenway scene was successfully obtained by applying the workflow. The model can accurately reproduce the spatial details of the greenway scene and the fitting degree is appropriate. Based on the two major scenarios of "weakly controlled" and "strongly controlled", with the aid of prompt words and the ControlNet, two effective paradigms for generation control are proposed: rapid generation of design intentions and redrawing of existing scenes. In addition, the use of various fine-tuning models can realize the tasks of generating error control and drawing style migration. Conclusion The workflow proposed in the research has some limitations. First, in terms of data collection, the image data from social platforms are not specialized enough, there are user preferences in the data, and the ownership of the dataset is unknown. Second, in terms of design-assisted generation, there is a lack of relevant mature models, a lack of specialization control methods for landscape gardens, and a poor interpretation of the generated results. In future research, new technological tools should be combined to gradually improve workflow performance and continuously reduce the cost of workflow deployment. However, the workflow also shows potential for adaptation in landscape architecture design, which is mainly reflected in two aspects: relevance and extensibility. The workflow provides a targeted path to real datasets based on real feedback from social media users on similar built cases. With the help of fine-tuned modeling techniques, it can be trained according to the specific needs of design practice. In addition, the workflow is scalable and can be quickly deployed for other tasks besides the greenway scenario practiced in the study.

     

/

返回文章
返回