variety of Richer Convolutional Features for Edge Detection (resnet101-based)
I test two types of loss: last layer loss / sum up each layer's loss. Detail here.
- last layer only: ODS: 0.8135 , OIS: 0.831 on BSDS500 dataset pretrained model, last layer only
- all layers: ODS: 0.796 , OIS: 0.814 on BSDS500 dataset pretrained model, all layers
- pytorch 0.4.1
- python 3.6.6
- dataset(provide by original repo)
- and other requirements... (cv2, numpy , etc.)
- put your data in 'data/HED-BSDS_PASCAL' (or make a soft link)
- python train.py
- python test.py
- python demo.py
it may take several hours...
- pretrained model, last layer only or pretrained model, all layers
- requirements:
- matlab
- hed link
- you should modify path to your predicts and path to ground truth (.mat)
- path to your predicts should contain two folders: png and mat
- sh eval.sh