JAIST Repository >
School of Knowledge Science >
Articles >
Journal Articles >

Please use this identifier to cite or link to this item: http://hdl.handle.net/10119/18792

Title: Sketch-Based Velocity Field Design using Latent Diffusion
Authors: Chang, Hengyuan
Peng, Yichen
Sato, Syuhei
Xie, Haoran
Keywords: Velocity field generation
latent diffusion model
Issue Date: 2023-09-09
Publisher: 情報処理学会
Magazine name: 研究報告コンピュータグラフィックスとビジュアル情報学(CG)
Volume: 2023-CG-191
Number: 8
Start page: 1
End page: 6
Abstract: The fusion of velocity field creation and deep learning to enhance the effectiveness of flow simulation is both widely pursued and intricate. Existing methods are mainly based on generative adversarial networks(GANs) to generate the target velocity field with the sketch as input. However, the training of GANs, and balancing the generator and discriminator is an unstable process, which causes generated unstable samples. In this research, we propose an interactive 2D velocity field design generation framework. In our framework, the streamline sketch is used as a constraint condition, maps into the latent space with an encoder, constrains the denoising process and reconstructs the 2D velocity field by decoder. The results show that our framework generates velocity fields corresponding to the shape of given sketches, and also allows users to reconstruct velocity fields from hand-drawn sketches. We compared our results with the GAN-based model. The evaluation shows our framework is more robust than the GAN-based method.
Rights: 社団法人情報処理学会, Hengyuan Chang, Yichen Peng, Syuhei Sato, Haoran Xie, 情報処理学会研究報告. CG, コンピュータグラフィックスとビジュアル情報学, 2023-CG-191 (8), 2023, pp.1-6. ここに掲載した著作物の利用に関する注意: 本著作物の著作権は(社)情報処理学会に帰属します。本著作物は著作権者である情報処理学会の許可のもとに掲載するものです。ご利用に当たっては「著作権法」ならびに「情報処理学会倫理綱領」に従うことをお願いいたします。 Notice for the use of this material: The copyright of this material is retained by the Information Processing Society of Japan (IPSJ). This material is published on this web site with the agreement of the author (s) and the IPSJ. Please be complied with Copyright Law of Japan and the Code of Ethics of the IPSJ if any users wish to reproduce, make derivative work, distribute or make available to the public any part or whole thereof. All Rights Reserved, Copyright (C) Information Processing Society of Japan.
URI: http://hdl.handle.net/10119/18792
Material Type: publisher
Appears in Collections:a10-1. 雑誌掲載論文 (Journal Articles)

Files in This Item:

File Description SizeFormat
H-XIE-K-1201-1.pdf4837KbAdobe PDFView/Open

All items in DSpace are protected by copyright, with all rights reserved.


Contact : Library Information Section, Japan Advanced Institute of Science and Technology