Open firasomran01 opened 5 years ago
Hi, Yes it would definitely help, you would need to load the ground truth depths when training and adding a regression loss on the non-zero depth points. Have a look at the new kitti benchmark for better ground truth data
Hi, Thank you for sharing your great work. I would like to use your method in a supervised fashion using the ground truth of Kitti dataset. what are the changes that i need to apply in order to train such a model. and do you think that this would give better results in terms of depth estimation ? Regards,