- [2022/10/24] We added more visualization results in another repository: BiCross-vis.
For more details, please refer to our paper on Arxiv.
python train.py
- Training stages:
- Since the pretrained parameters of DPT are trained on the ImageNet, when you train from scratch, please first pretrain the model on the source RGB to adapt to the depth estimation task,, changing the
stage
option in thetrain_config.json
topretrain
and training for about 30 epochs. - After the pretrain stage, set
stage
intrain_config.json
tocrossmodality
and continue training for another 10 epochs from source RGB to source spike. - Finally, set
stage
intrain_config.json
tocrossdomain
and then continue training for about 20 epochs from source spike to target spike.
- Since the pretrained parameters of DPT are trained on the ImageNet, when you train from scratch, please first pretrain the model on the source RGB to adapt to the depth estimation task,, changing the
python test.py
python visualize.py
You can modify the configs for different training and testing configurations.
Coming soon (in Google Drive) !