Mengyang Pu

https://github.com/MengyangPu/RINDNet/blob/main/evaluate.py

https://github.com/MengyangPu/RINDNet/blob/main/evaluate.py or https://github.com/MengyangPu/RINDNet/blob/main/evaluate_edge.py

Hello, we did not fix the seed during training, resulting in different random numbers.

Please try to change the size in config files, e.g., configs/bsds/EDTER_BIMLA_320x320_80k_bsds_local8x8_bs_8.py or configs/_base_/models/edter_bimla_local8x8.py.

https://pan.baidu.com/s/1_daeYDxCrKNGSU0cB1v8jw?pwd=nsxh

Sorry, the capacity of our Google Drive is limited. If you cannot download from BaiduDisk, could you provide me with an upload method? We will upload the files.

Please try the following Matlab code: ` result = load('your path of the .mat file'); imwrite(result,'saving path'); `

Sorry, the capacity of our Google Drive is limited. If you cannot download from BaiduDisk, could you provide me with an upload method? We will upload the files.

Thanks, this fix will not affect training, or replacing line with: `loss = self.criterion(torch.cat((out_depth,out_normal,out_reflectance,out_illumination),dim=1), target[:, 1:5, :, :])`

Sorry, [depth_decoder.py](https://github.com/MengyangPu/RINDNet/blob/main/modeling/depth_networks/depth_decoder.py) is not used, please ignore this error. Thanks~