Software Open Access

Deeplabv3+ models for the aradeepopsis pipeline

Patrick Hüther; Niklas Schandry; Claude Becker

Transfer learning of the DeepLabv3+ model (xception backbone) for use with aradeepopsis in single- or multi-scale mode.

1_class corresponds to model A; 2_class to model B and 3_class to model C as described here

The datasets used for training are available at https://doi.org/10.5281/zenodo.3946393 and the code used for transfer learning is available on GitHub

Files (1.0 GB)
Name Size
1_class_multiscale.pb
md5:52eb51e5b62199d0e8167d28f171c990
169.1 MB Download
1_class_singlescale.pb
md5:bfe74d476b5f9dee87c8af9503d2d2fe
165.7 MB Download
2_class_multiscale.pb
md5:cecd65a73c3090d3b4bcd75c6b88f416
169.1 MB Download
2_class_singlescale.pb
md5:add1e6dc8df75261cd0cefc9d8c398dc
165.7 MB Download
3_class_multiscale.pb
md5:e3c87022467c568bf35b50c2295b26ba
169.1 MB Download
3_class_singlescale.pb
md5:f308e6a748ee587af1eb31ce968cf735
165.7 MB Download
  • Hüther, P.; Schandry, N.; Jandrasits, K.; Bezrukov, I.; Becker, C. aradeepopsis: From images to phenotypic traits using deep transfer learning. bioRxiv, 2020, 2020.04.01.018192.

  • Chen, L.-C.; Zhu, Y.; Papandreou, G.; Schroff, F.; Adam, H. Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. arXiv [cs.CV], 2018.

163
552
views
downloads
All versions This version
Views 163163
Downloads 552552
Data volume 92.0 GB92.0 GB
Unique views 134134
Unique downloads 231231

Share

Cite as