How Will It Drape Like? Capturing Fabric Mechanics from Depth Images
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866929274374586368 |
|---|---|
| author | Rodriguez-Pardo, Carlos Prieto-Martin, Melania Casas, Dan Garces, Elena |
| author_facet | Rodriguez-Pardo, Carlos Prieto-Martin, Melania Casas, Dan Garces, Elena |
| contents | We propose a method to estimate the mechanical parameters of fabrics using a casual capture setup with a depth camera. Our approach enables to create mechanically-correct digital representations of real-world textile materials, which is a fundamental step for many interactive design and engineering applications. As opposed to existing capture methods, which typically require expensive setups, video sequences, or manual intervention, our solution can capture at scale, is agnostic to the optical appearance of the textile, and facilitates fabric arrangement by non-expert operators. To this end, we propose a sim-to-real strategy to train a learning-based framework that can take as input one or multiple images and outputs a full set of mechanical parameters. Thanks to carefully designed data augmentation and transfer learning protocols, our solution generalizes to real images despite being trained only on synthetic data, hence successfully closing the sim-to-real loop.Key in our work is to demonstrate that evaluating the regression accuracy based on the similarity at parameter space leads to an inaccurate distances that do not match the human perception. To overcome this, we propose a novel metric for fabric drape similarity that operates on the image domain instead on the parameter space, allowing us to evaluate our estimation within the context of a similarity rank. We show that out metric correlates with human judgments about the perception of drape similarity, and that our model predictions produce perceptually accurate results compared to the ground truth parameters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2304_06704 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | How Will It Drape Like? Capturing Fabric Mechanics from Depth Images Rodriguez-Pardo, Carlos Prieto-Martin, Melania Casas, Dan Garces, Elena Computer Vision and Pattern Recognition Artificial Intelligence Computational Geometry Graphics Machine Learning 68T45 (Primary) 68U05 (Secondary) I.5.4; I.3.5; I.4.8; I.5.1 We propose a method to estimate the mechanical parameters of fabrics using a casual capture setup with a depth camera. Our approach enables to create mechanically-correct digital representations of real-world textile materials, which is a fundamental step for many interactive design and engineering applications. As opposed to existing capture methods, which typically require expensive setups, video sequences, or manual intervention, our solution can capture at scale, is agnostic to the optical appearance of the textile, and facilitates fabric arrangement by non-expert operators. To this end, we propose a sim-to-real strategy to train a learning-based framework that can take as input one or multiple images and outputs a full set of mechanical parameters. Thanks to carefully designed data augmentation and transfer learning protocols, our solution generalizes to real images despite being trained only on synthetic data, hence successfully closing the sim-to-real loop.Key in our work is to demonstrate that evaluating the regression accuracy based on the similarity at parameter space leads to an inaccurate distances that do not match the human perception. To overcome this, we propose a novel metric for fabric drape similarity that operates on the image domain instead on the parameter space, allowing us to evaluate our estimation within the context of a similarity rank. We show that out metric correlates with human judgments about the perception of drape similarity, and that our model predictions produce perceptually accurate results compared to the ground truth parameters. |
| title | How Will It Drape Like? Capturing Fabric Mechanics from Depth Images |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Computational Geometry Graphics Machine Learning 68T45 (Primary) 68U05 (Secondary) I.5.4; I.3.5; I.4.8; I.5.1 |
| url | https://arxiv.org/abs/2304.06704 |