@inproceedings{koerschens2021weakly,
    type = {inproceedings},
    key = {koerschens2021weakly},
    title = {Weakly Supervised Segmentation Pretraining for Plant Cover Prediction},
    author = {Matthias Körschens and Paul Bodesheim and Christine Römermann and Solveig Franziska Bucher and Mirco Migliavacca and Josephine Ulrich and Joachim Denzler},
    booktitle = {DAGM German Conference on Pattern Recognition (DAGM-GCPR)},
    year = {2021},
    pages = {589-603},
    abstract = {Automated plant cover prediction can be a valuable tool for botanists, as plant cover estimations are a laborious and recurring task in environmental research. Upon examination of the images usually encompassed in this task, it becomes apparent that the task is ill-posed and successful training on such images alone without external data is nearly impossible. While a previous approach includes pretraining on a domain-related dataset containing plants in natural settings, we argue that regular classification training on such data is insufficient. To solve this problem, we propose a novel pretraining pipeline utilizing weakly supervised object localization on images with only class annotations to generate segmentation maps that can be exploited for a second pretraining step. We utilize different pooling methods during classification pretraining, and evaluate and compare their effects on the plant cover prediction. For this evaluation, we focus primarily on the visible parts of the plants. To this end, contrary to previous works, we created a small dataset containing segmentations of plant cover images to be able to evaluate the benefit of our method numerically. We find that our segmentation pretraining approach outperforms classification pretraining and especially aids in the recognition of less prevalent plants in the plant cover dataset.},
    doi = {10.1007/978-3-030-92659-5_38},
    groups = {biodiversity},
}