search

ShuweiShao / AF-SfMLearner

[MedIA2022 & ICRA2021] Self-Supervised Monocular Depth and Ego-Motion Estimation in Endoscopy: Appearance Flow to the Rescue
MIT License
101 stars 15 forks source link

prepare the ground truth depth maps #57

Open wwj-53 opened 3 months ago

wwj-53 commented 3 months ago

Based on your description，the directory of dataset structure is shown as ： /path/to/endovis_data/ dataset1/ keyframe1/ image_02/ data/ 0000000001.png

but after processing according to your method, it looks like this: /path/to/endovis_data/ dataset1/ keyframe1/ data/ 0000000001.png There is no image folder name available.

When I run CUDA_VISIBLE_DEVICES=0 python export_gt_depth.py --data_path /media/wwj/aj_new/dataset_2 --split endovis， The error is as follows： (endodac) wwj@cpnt:~/3DGS/AF-SfMLearner$ CUDA_VISIBLE_DEVICES=0 python export_gt_depth.py --data_path /media/wwj/aj_new/dataset2 --split endovis Exporting ground truth depths for endovis 1 dataset3/keyframe4 [ WARN:0@0.005] global loadsave.cpp:241 findDecoder imread('/media/wwj/aj_new/dataset_2/dataset3/keyframe4/data/groundtruth/scene_points000389.tiff'): can't open/read file: check file path/integrity Traceback (most recent call last): File "/home/wwj/3DGS/AF-SfMLearner/export_gt_depth.py", line 61, in export_gt_depths_SCARED() File "/home/wwj/3DGS/AF-SfMLearner/export_gt_depth.py", line 48, in export_gt_depths_SCARED depth_gt = depth_gt[:, :, 0] TypeError: 'NoneType' object is not subscriptable

I have changed line 44 from "image-02/data/roundtruth" to "data/roundtruth" (otherwise it will not run according to the structure of my dataset)

suxuanya commented 3 months ago

I also encountered this problem. Have you solved it?

wwj-53 commented 3 months ago

I also encountered this problem. Have you solved it?

no

ShuweiShao commented 3 months ago

Hi all,

Please create the image_02 folder by yourself and put the data in it. The format follows the KITTI dataset and ''image_02'' indicates the left frames.

suxuanya commented 3 months ago

Thanks a lot.