search

lliuz / ARFlow

The official PyTorch implementation of the paper "Learning by Analogy: Reliable Supervision from Transformations for Unsupervised Optical Flow Estimation".
MIT License
251 stars 50 forks source link

KIITI datasets used for training and finetuning #28

Open RokiaAbdeen opened 3 years ago

RokiaAbdeen commented 3 years ago

I downloaded the following files : Kitti raw , data_stereo_flow for kitti 2012 , data_scene_flow for kitti 2015 and data_stereo_flow_multiview , data_scene_flow_multiview first I want to train the model with kittiraw and validate using kitti stereo and kitti scene but I got data assertion error , I thinks that's because may be there is wrong dataset used could you explain please "according to your config files " what should be the files specified for the pretraining and finetuning if I want to pretrain with kitti raw and fine tune with multiview as you explained in your paper

thanks in advance ..

RokiaAbdeen commented 3 years ago

how did you finetune by multi-view dataset ? in finetuning you specified kitti15 as the training root not multiview file, so the training data here is same to the validation data , that's not same to paper , could you explain this point for me please when I tried to use multiview dataset for filetuning I got data assertion error because there is no flow_occ folder in multiview dataset thanks in advance ..