Huangying-Zhan / Depth-VO-Feat

Unsupervised Learning of Monocular Depth Estimation and Visual Odometry with Deep Feature Reconstruction
Other
349 stars 66 forks source link

when run ./experiments/depth_odometry_feature/train.sh #8

Closed WZYSilverWing closed 6 years ago

WZYSilverWing commented 6 years ago

in ./experiments/depth_odometry_feature/train.sh, it requires Weerasekera_nyu.caffemodel, where can I dowload this caffe model? thank you!

Huangying-Zhan commented 6 years ago

Hi @WZYSilverWing , the model is not yet released. The model is trained from this paper, Weerasekera et al.. I will post a link to the model later. If you want to use the feature construction loss, please try the KITTI_feat.caffemodel for now. Sorry for the inconvenience caused.

WZYSilverWing commented 6 years ago

thank you so much!