ControlNet-based domain adaptation for synthetic construction images via graphical simulation and generative AI
Authors:
- Sina DavariSDSina Davari
- Daeho KimDKDaeho Kim
- Ali TohidifarATAli Tohidifar
Open Access
- 30 September 2025
- journal article
- Published by Elsevier in Automation in Construction
- Vol. 180
Abstract
No abstract availableFunding Information
- Mitacs (IT403)
- NSERC (RGPIN-2022-04429, ALLRP 597479-24)
This publication has 23 references indexed in Scilit:
- Make it till you fake it: Construction-centric computational framework for simultaneous image synthetization and multimodal labelingAutomation in Construction, 2024
- A review of convolutional neural networks in computer visionArtificial Intelligence Review, 2024
- Synthetic image data generation using BIM and computer graphics for building scene understandingAutomation in Construction, 2023
- Hybrid DNN training using both synthetic and real construction images to overcome training data shortageAutomation in Construction, 2023
- SODA: A large-scale open site object detection dataset for deep learning in constructionAutomation in Construction, 2022
- Deep learning-based data analytics for safety in constructionAutomation in Construction, 2022
- Imperfect ImaGANation: Implications of GANs exacerbating biases on facial data augmentation and snapchat face lensesArtificial Intelligence, 2022
- Synthetic data generation using building information modelsAutomation in Construction, 2021
- Dataset and benchmark for detecting moving objects in construction sitesAutomation in Construction, 2021
- Deep neural networks architecture driven by problem-specific informationNeural Computing and Applications, 2021