Abstract
For computer vision systems based on artificial neural networks, increasing the resolution of images typically improves the performance of the network. However, ImageNet pre-trained Vision Transformer (ViT) models are typically only openly available for 2242 and 3842 image resolutions. To determine the impact of using higher resolution images with ViT systems the performance differences between ViT-B/16 models (designed for 3842 and 5442 image resolutions) were evaluated. The multi-label classification RANZCR CLiP challenge dataset, which contains over 30,000 high resolution labelled chest X-ray images, was used throughout this investigation. The performance of the ViT 3842 and ViT 5442 models with no ImageNet pre-training (i.e. models were only trained using RANZCR data) was firstly compared to see if using higher resolution images increases performance. After this, a multi-resolution fine-tuning approach was investigated for transfer learning. This approach was achieved by transferring learned parameters from ImageNet pre-trained ViT 3842 models, which had undergone further training on the 3842 RANZCR data, to ViT 5442 models which were then trained on the 5442 RANZCR data. Learned parameters were transferred via a tensor slice copying technique. The results obtained provide evidence that using larger image resolutions positively impacts ViT network performance and that multi-resolution fine-tuning can lead to performance gains. The multi-resolution fine-tuning approach used in this investigation could potentially improve the performance of other computer vision systems which use ViT based networks. The results of this investigation may also warrant the development of new ViT variants optimized to work with high resolution image datasets.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Tan, M., Le, Q.: EfficientNet: rethinking model scaling for convolutional neural networks. In: International Conference on Machine Learning (ICML), Long Beach (2019)
Tan, M., Le, Q.: EfficientNetV2: smaller models and faster training. In: International Conference on Machine Learning (ICML), Virtual (2021)
Deng, J., Dong, W., Socher, R., Li, L., Li, K., Fei-Fei, L.: ImageNet: a large-scale hierarchical image database. In: IEEE Conference on Computer Vision and Pattern Recognition, Miami (2009)
Touvron, H., Vedaldi, A., Douze, M., Jegou, H.: Fixing the train-test resolution discrepancy. In: Advances in Neural Information Processing Systems (2019)
Kolesnikov, A., et al.: Big transfer (BiT): general visual representation learning. In: Vedaldi, A., Bischof, H., Brox, T., Frahm, J.-M. (eds.) ECCV 2020. LNCS, vol. 12350, pp. 491–507. Springer, Cham (2020). https://doi.org/10.1007/978-3-030-58558-7_29
Dosovitskiy, A., et al.: An image is worth 16x16 words: transformers for image recognition at scale. In: International Conference on Learning Representations (ICLR), Virtual (2021)
Zhai, X., Kolesnikov, A., Houlsby, N., Beyer, L.: Scaling vision transformers. arXiv (2021)
Tang, J., et al.: CLiP, catheter and line position dataset. Sci. Data 8, 1–7 (2021)
Law, M., et al.: RANZCR CLiP - catheter and line position challenge. Kaggle, 8 March 2021. https://www.kaggle.com/competitions/ranzcr-clip-catheter-line-classification/overview. Accessed June 2021
Wightman, R.: PyTorch image models. GitHub Repository (2019)
Ranftl, R., Bochkovskiy, A., Koltun, V.: Vision transformers for dense prediction. In: International Conference on Computer Vision (ICCV), Montreal (2021)
Silberman, N., Hoiem, D., Kohli, P., Fergus, R.: Indoor segmentation and support inference from RGBD images. In: Fitzgibbon, A., Lazebnik, S., Perona, P., Sato, Y., Schmid, C. (eds.) ECCV 2012. LNCS, vol. 7576, pp. 746–760. Springer, Heidelberg (2012). https://doi.org/10.1007/978-3-642-33715-4_54
Zhang, P., et al.: Multi-scale vision longformer: a new vision transformer for high-resolution image encoding. In: International Conference on Computer Vision (ICCV), Virtual (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Switzerland AG
About this paper
Cite this paper
Fitzgerald, K., Law, M., Seah, J., Tang, J., Matuszewski, B. (2022). Multi-resolution Fine-Tuning of Vision Transformers. In: Yang, G., Aviles-Rivero, A., Roberts, M., Schönlieb, CB. (eds) Medical Image Understanding and Analysis. MIUA 2022. Lecture Notes in Computer Science, vol 13413. Springer, Cham. https://doi.org/10.1007/978-3-031-12053-4_40
Download citation
DOI: https://doi.org/10.1007/978-3-031-12053-4_40
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-12052-7
Online ISBN: 978-3-031-12053-4
eBook Packages: Computer ScienceComputer Science (R0)