search

mli0603 / stereo-transformer

Revisiting Stereo Depth Estimation From a Sequence-to-Sequence Perspective with Transformers. (ICCV 2021 Oral)
Apache License 2.0
660 stars 107 forks source link

Training on KITTI #84

Open suriya030 opened 1 year ago

suriya030 commented 1 year ago

I want to train the network on kitti dataset , as I'm not much experienced. Could you please point out what changes should I have to do in the python scripts along with where do I have to load the dataset ? Thankyou in advance .

18370271993 commented 1 year ago

I have the same problem

mli0603 commented 1 year ago

Hi @suriya030 and @18370271993

Sorry for the late response. Was the instruction in this link not helpful?

The folder structure should be intact after downloading the KITTI data as follows:

KITTI_2015
    |_ training
        |_ disp_occ_0 (disparity including occluded region)
        |_ image_2 (left image)
        |_ image_3 (right image)

And what you need to do is providing the path to kitti folder in the line below: https://github.com/mli0603/stereo-transformer/blob/d0aa1ad9c84f3dab15a2f2a9ead2ca6cf9fe8971/scripts/kitti_finetune.sh#L8

WIFIwifi8966 commented 2 months ago

I want to train the network on kitti dataset , as I'm not much experienced. Could you please point out what changes should I have to do in the python scripts along with where do I have to load the dataset ? Thankyou in advance .

hello I have the same question with you. Did you finish this work? If possible I want to pay for you for asking some questions about this