Balanced sampling for an object detection problem - application to fetal anatomies detection

Antoine Olivier, Caroline Raynaud
Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, PMLR 143:554-566, 2021.

Abstract

In this paper, we propose a novel approach to overcome the problem of imbalanced datasets for object detection tasks, when the distribution is not uniform over all classes. The general idea is to compute a probability vector, encoding the probability for each image to be fed to the network during the training phase. This probability vector is computed by solving some quadratic optimization problem and ensures that all classes are seen with similar frequency. We apply this method to a fetal anatomies detection problem, and conduct a thorough statistical analysis of the resulting performance to show that it performs significantly better than two baseline models: one with images sampled uniformly and one implementing classical oversampling.

Cite this Paper

BibTeX
@InProceedings{pmlr-v143-olivier21a, title = {Balanced sampling for an object detection problem - application to fetal anatomies detection}, author = {Olivier, Antoine and Raynaud, Caroline}, booktitle = {Proceedings of the Fourth Conference on Medical Imaging with Deep Learning}, pages = {554--566}, year = {2021}, editor = {Heinrich, Mattias and Dou, Qi and de Bruijne, Marleen and Lellmann, Jan and Schläfer, Alexander and Ernst, Floris}, volume = {143}, series = {Proceedings of Machine Learning Research}, month = {07--09 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v143/olivier21a/olivier21a.pdf}, url = {https://proceedings.mlr.press/v143/olivier21a.html}, abstract = {In this paper, we propose a novel approach to overcome the problem of imbalanced datasets for object detection tasks, when the distribution is not uniform over all classes. The general idea is to compute a probability vector, encoding the probability for each image to be fed to the network during the training phase. This probability vector is computed by solving some quadratic optimization problem and ensures that all classes are seen with similar frequency. We apply this method to a fetal anatomies detection problem, and conduct a thorough statistical analysis of the resulting performance to show that it performs significantly better than two baseline models: one with images sampled uniformly and one implementing classical oversampling.} }
Endnote
%0 Conference Paper %T Balanced sampling for an object detection problem - application to fetal anatomies detection %A Antoine Olivier %A Caroline Raynaud %B Proceedings of the Fourth Conference on Medical Imaging with Deep Learning %C Proceedings of Machine Learning Research %D 2021 %E Mattias Heinrich %E Qi Dou %E Marleen de Bruijne %E Jan Lellmann %E Alexander Schläfer %E Floris Ernst %F pmlr-v143-olivier21a %I PMLR %P 554--566 %U https://proceedings.mlr.press/v143/olivier21a.html %V 143 %X In this paper, we propose a novel approach to overcome the problem of imbalanced datasets for object detection tasks, when the distribution is not uniform over all classes. The general idea is to compute a probability vector, encoding the probability for each image to be fed to the network during the training phase. This probability vector is computed by solving some quadratic optimization problem and ensures that all classes are seen with similar frequency. We apply this method to a fetal anatomies detection problem, and conduct a thorough statistical analysis of the resulting performance to show that it performs significantly better than two baseline models: one with images sampled uniformly and one implementing classical oversampling.
APA
Olivier, A. & Raynaud, C.. (2021). Balanced sampling for an object detection problem - application to fetal anatomies detection. Proceedings of the Fourth Conference on Medical Imaging with Deep Learning, in Proceedings of Machine Learning Research 143:554-566 Available from https://proceedings.mlr.press/v143/olivier21a.html.

Related Material