train on own dataset with fm_joint.cfg

sconlyshootery / FeatDepth

This is the offical codes for the methods described in the "Feature-metric Loss for Self-supervised Learning of Depth and Egomotion".

MIT License

247 stars 28 forks source link

train on own dataset with fm_joint.cfg #83

Closed ThomasR155 closed 2 years ago

ThomasR155 commented 2 years ago

I would like to train the mono_fm_joint model on my own dataset. Should I revise the folder_dataset.py or the kitti_dataset.py (mono_dataset.py)?

sconlyshootery commented 2 years ago

you may feel more convenient to modify folder_dataset.py.