Comments (4)
Hi Shreyas,
The KITTI depth completion result is from a follow-up paper, and the code will be released later in another repo. For clarity, the "KITTI" dataset used in this repo is actually the KITTI odometry dataset, not the KITTI depth dataset (which had not been released when we started out this project).
from sparse-to-dense.pytorch.
Ahh, I see. However, can I assume that to get the results mentioned in https://arxiv.org/pdf/1709.07492.pdf I can use the parameters mentioned in the paper / default parameters in this repo?
from sparse-to-dense.pytorch.
Short answer: yes.
Longer answer: The results in the paper were generated using the Torch implementation, and I haven't done a detailed benchmarking for this pytorch code yet. However, I believe they should be pretty well in line with each other (maybe with some minor differences in accuracy).
from sparse-to-dense.pytorch.
Great, thanks for the information. I appreciate the quick responses!
Good luck on publishing your follow up work.
from sparse-to-dense.pytorch.
Related Issues (20)
- An issue with "resume" mode HOT 2
- [NYU] Different Scaling in Training and Validation HOT 4
- Implementing SLAM
- How is the loss calculated for KITTI dataset ?
- Is there an easy way to run inference on a different dataset HOT 2
- Apply the pretrained model to other datasets HOT 2
- No rgb image normalization during pre-process HOT 1
- Different sparse input when each sample input is loaded HOT 3
- The low download speed in NYU and KITTI
- License for repo
- pose information for processed data
- Benchmark on KITTI vs NYU Depth v2
- Request for pretrained model with depth-only modality
- Failed to reproduce the RGB based problem, whereas the RGBd problem works fine for me.
- How can I use this Git from Windows OS HOT 1
- Using another model
- Scaling factor cancels out for depth values
- The principle of implementing a simple Visual Odometry (VO) algorithm
- Output for custom image
- replace the method of "misc.imresize(img, self.size, self.interpolation, 'F')" HOT 2
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 sparse-to-dense.pytorch.