Abstract
How to build machine learning models from few annotations is an open research question. This article shows an application of a meta-learning algorithm (REPTILE) to solve the problem of object segmentation. We evaluate how using REPTILE during a pre-training phase accelerates the learning process without loosing performance of the resulting segmentation in poor labeling conditions, and compare these results against training the detectors using basic transfer learning. Two scenarios are tested: (i) how segmentation performance evolves through training epochs with a fixed amount of labels and (ii) how segmentation performance improves with an increasing amount of labels after a fixed amount of epochs. The results suggest that REPTILE is useful making learning faster in both cases.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
In aerial images it translates into different image ground resolutions and object scales.
- 2.
- 3.
Not anymore, given the increasing availability of electronic devices capable of capturing images (e.g., smartphones, satellites) nowadays.
- 4.
There are sub-datasets with diverse images of the same type at different resolutions.
- 5.
Annotation may be very time consuming, and that is the reason why we only tested one dataset.
References
Caruana, R.: Multitask learning. In: Thrun, S., Pratt, L. (eds.) Learning to Learn, pp. 95–133. Springer, Boston (1998). https://doi.org/10.1007/978-1-4615-5529-2_5
Darlow, L.N., Crowley, E.J., Antoniou, A., Storkey, A.J.: CINIC-10 is not ImageNet or CIFAR-10 (2018)
Paszke, A., et al.: ENet: a deep neural network architecture for real-time semantic segmentation. CoRR, abs/1606.02147 (2016)
Pang, J., et al.: Libra R-CNN: towards balanced learning for object detection. CoRR, abs/1904.02701 (2019)
Redmon, J., et al.: You only look once: unified, real-time object detection. CoRR, abs/1506.02640 (2015)
He, K., et al.: Spatial pyramid pooling in deep convolutional networks for visual recognition. CoRR, abs/1406.4729 (2014)
He, K., et al.: Mask R-CNN. CoRR, abs/1703.06870 (2017)
Russakovsky, O., et al.: ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. (IJCV) 115(3), 211–252 (2015). https://doi.org/10.1007/s11263-015-0816-y
Girshick, R.B., et al.: Rich feature hierarchies for accurate object detection and semantic segmentation. CoRR, abs/1311.2524 (2013)
Ren, S., et al.: Faster R-CNN: towards real-time object detection with region proposal networks. CoRR, abs/1506.01497 (2015)
Lin, T.-Y., et al.: Microsoft COCO: common objects in context. In: Fleet, D., Pajdla, T., Schiele, B., Tuytelaars, T. (eds.) ECCV 2014. LNCS, vol. 8693, pp. 740–755. Springer, Cham (2014). https://doi.org/10.1007/978-3-319-10602-1_48
Liu, W., et al.: SSD: single shot multibox detector. CoRR, abs/1512.02325 (2015)
Zhou, Z., Rahman Siddiquee, M.M., Tajbakhsh, N., Liang, J.: UNet++: a nested U-Net architecture for medical image segmentation. In: Stoyanov, D., et al. (eds.) DLMIA/ML-CDS -2018. LNCS, vol. 11045, pp. 3–11. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-00889-5_1
Finn, C., Abbeel, P., Levine, S.: Model-agnostic meta-learning for fast adaptation of deep networks (2017)
Girshick, R.: Fast R-CNN (2015)
Kamrul Hasan, S.M., Linte, C.A.: U-NetPlus: a modified encoder-decoder u-net architecture for semantic and instance segmentation of surgical instrument. CoRR, abs/1902.08994 (2019)
Hodgson, J., et al.: Drones count wildlife more accurately and precisely than humans. Methods Ecol. Evol. 9, 1160–1167 (2018)
Krizhevsky, A., Nair, V., Hinton, G.: CIFAR-10 (Canadian Institute for Advanced Research)
Nichol, A., Achiam, J., Schulman, J.: On first-order meta-learning algorithms. CoRR, abs/1803.02999 (2018)
Redmon, J., Farhadi, A.: YOLO9000: better, faster, stronger (2016)
Redmon, J., Farhadi, A.: Yolov3: an incremental improvement (2018)
Ronneberger, O., Fischer, P., Brox, T.: U-Net: convolutional networks for biomedical image segmentation. CoRR, abs/1505.04597 (2015)
Vanschoren, J.: Meta-learning: a survey (2018)
Wang, Y., Yao, Q., Kwok, J., Ni, L.M.: Generalizing from a few examples: a survey on few-shot learning (2019)
Zeng, Z., Xie, W., Zhang, Y., Lu, Y.: RIC-Unet: an improved neural network based on Unet for nuclei segmentation in histology images. IEEE Access 7, 21420–21428 (2019)
Acknowledgments
This work was supported by the Swiss Space Center (SERI/SSO MdP program). All the experiments shown in this paper were performed thanks to a thight collaboration with the company Picterra (https://picterra.ch/). Picterra provided all the datasets used, and participated in constructive discussion about how to deal with large images and how to build object detectors from few examples.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2021 Springer Nature Switzerland AG
About this paper
Cite this paper
Satizábal, H.F., Perez-Uribe, A. (2021). Learning Image Segmentation from Few Annotations. In: Rojas, I., Joya, G., Català, A. (eds) Advances in Computational Intelligence. IWANN 2021. Lecture Notes in Computer Science(), vol 12861. Springer, Cham. https://doi.org/10.1007/978-3-030-85030-2_42
Download citation
DOI: https://doi.org/10.1007/978-3-030-85030-2_42
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-85029-6
Online ISBN: 978-3-030-85030-2
eBook Packages: Computer ScienceComputer Science (R0)