GreenCanvas: 3D Model Plant Generator System Based on Drawings
| datacite.subject.fos | Engenharia e Tecnologia::Engenharia Eletrotécnica, Eletrónica e Informática | |
| dc.contributor.advisor | Gonçalves, Alexandrino José Marques | |
| dc.contributor.advisor | Marto, Anabela Gonçalves Rodrigues | |
| dc.contributor.advisor | Ribeiro, Roberto Aguiar | |
| dc.contributor.advisor | Ascenso, Rita Margarida Teixeira | |
| dc.contributor.author | Martins, Ana Rita Marques da Silva | |
| dc.date.accessioned | 2025-12-17T09:31:03Z | |
| dc.date.available | 2025-12-17T09:31:03Z | |
| dc.date.issued | 2025-11-10 | |
| dc.description.abstract | The generation of three-dimensional plant models is increasingly important for applications in agriculture, scientific visualization, landscape design, and immersive virtual environments. Existing approaches often rely on specialized hardware or sensor data, and publicly available models are limited, repetitive, or constrained to specific species, limiting creative flexibility. This dissertation addresses these challenges by focusing on sketch-based plant modelling as an accessible and flexible approach. A systematic review of state-of-the-art methods was conducted following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) guidelines, identifying two main directions: sketch-based modelling and tree generation from photographic data. Insights from this review informed the design of a novel framework that integrates single- and multi-sketch strategies to generate 3D plant models from user drawings. The framework was implemented as a Blender add-on, supporting features such as variable levels of detail, leaf propagation, and generation of multiple model variations, while providing a streamlined and intuitive interface. The add-on was evaluated through quantitative and qualitative analyses. Task completion times demonstrated efficient modelling, and a mean System Usability Scale (SUS) score of 56.4 indicated moderate usability. Correlations with demographic factors revealed influences of user experience, background, and familiarity with input devices on usability. Qualitative feedback highlighted areas for improvement in feature intuitiveness and onboarding. Despite a limited and unevenly distributed participant sample, the results indicate that the framework is a promising tool for procedural plant modelling and can enhance creative workflows for users of varying expertise. This work contributes both a systematic understanding of visual-based plant modelling approaches and a practical tool that bridges sketching and 3D model creation, providing a foundation for future enhancements such as AI-assisted sketch interpretation, expanded plant element support, and improved interface features. | por |
| dc.identifier.tid | 204063973 | |
| dc.identifier.uri | http://hdl.handle.net/10400.8/15114 | |
| dc.language.iso | eng | |
| dc.relation | Research Center in Informatics and Communications | |
| dc.rights.uri | http://creativecommons.org/licenses/by/4.0/ | |
| dc.subject | Plant | |
| dc.subject | 3D Plant Modelling | |
| dc.subject | Blender add-on | |
| dc.subject | 2D to 3D | |
| dc.subject | Sketch to 3D | |
| dc.subject | 3D Generation | |
| dc.title | GreenCanvas: 3D Model Plant Generator System Based on Drawings | |
| dc.type | master thesis |