Comments (5)
And is the preprocess necessary for the work of training? When I run the "python trainer.py", it seems another document is needed:
'/data/Kitti_RAW_Data/2011_09_30/2011_09_30_drive_0028_sync/2channel/5001_r_False.npy'
I'm puzzeled about how to get these data.
from fusiondepth.
Hi Renatusphere:
Thank you for your interest in our work, yes this preprocess is necessary for the training.
The file '4beam/xxx.bin' is generated by the sparsify.py, which samples 4 LiDAR lines from the KITTI original 64-beam data. I think the problem is that the output path here does not match your data path (/data/Kitti_RAW_Data/...).
You can either soft link your data path to the FusionDepth root folder as "kitti_data" or feed it to the sparsify.py.
I hope this will solve your problem, thank you again for your interest!
from fusiondepth.
Thank you for your response!
It seems that the error happens in the process of loading the velodyne data. In the code below, it tries to load raw data from velo_filename: "/data/Kitti_RAW_Data/2011_09_26/2011_09_26_drive_0002_sync/4beam/0000000069.bin"
However, I assume that this file should have been the output file, so the velo_filename is incorrect.
I tried to modify my output path, but it didn't work. It turned out to be the same error.
Should I modify the code of splicing velo_filename? Thank you again!
from fusiondepth.
The raw 64-beam Velodyne data is 'velodyne_points/data/xxx.bin' while the sparsified data is '4beam/xxx.bin'. The variable name "velo_filename" here is not clear enough, you are welcome to modify it and submit a pull request.
Could you check where your sparsify.py outputted the '4beam/xxx.bin' files? that's what the code in your screenshot supposed to load.
from fusiondepth.
It turned out that I missed a "/" in the end of ptc_path, which lead to no output.
Thanks a lot for your great work and enthusiastic response!
from fusiondepth.
Related Issues (20)
- No such file or directory HOT 4
- Absolute depth or relative depth? HOT 2
- Error when run evaluation by pretrained model HOT 3
- Running trainer.py failed运行trainer.py失败 HOT 7
- 运行refiner.py时loss为nan HOT 7
- 请问为什么论文里Initial Depth结果与预训练模型结果不一样? HOT 2
- Error when run evaluation on training result HOT 6
- 同样的权重文件训练过程中val的结果与evaluate_depth.py运行的结果不一致? HOT 3
- Accurate or normalized depth value in training? HOT 4
- Visualization HOT 1
- prepare_4beam_data_for_prediction error
- the quantitative comparsion in your paper
- Can you share the pretrained model HOT 1
- About the depth completion task HOT 2
- The evaluation_depth.py is missing HOT 3
- about static scenario HOT 2
- No ptc2depth ! HOT 5
- questions about detection HOT 1
- Tensor size matching error HOT 7
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from fusiondepth.