Weakly-Supervised Crack Detection Dataset

This repo contains two files: crack detection dataset (weakly_sup_crackdet_dataset.zip), and pretrained TensorFlow model for Xception65 (pascal_voc_seg.zip).

The dataset consists of rough annotations used in weakly-supervised crack detection. It contains roughly annotated ground truths for the following datasets:

• Aigle

• Crack Forest Dataset

• DeepCrack

Annotations of different "roughness" are stored. Directories suffixed "*_dil*" are synthetically-generated annotations, while directories suffixed "*_rough" and "*_rougher" are manually-generated annotations. The detail of the dataset is described in [1]. Please also refer to our GitHub repo https://github.com/hitachi-rd-cv/weakly-sup-crackdet for more details.

This dataset is made available by Hitachi, Ltd.

The pretrained model is used by [1]. Please use it for comparison experiments. Please refer to our GitHub repor for more details.

[1] Inoue, Y., Nagayoshi, H.: Crack detection as a weakly-supervised problem: Towards achieving less annotation-intensive crack detectors. In: International Conference on Pattern Recognition (ICPR) (2020)