A Pytorch implementation(unofficial) for paper "PackNet-SfM: 3D Packing for Self-Supervised Monocular Depth Estimation"
License: GNU General Public License v3.0
This codebase almost implements(unofficial) the system described in the paper:
PackNet-SfM: 3D Packing for Self-Supervised Monocular Depth Estimation
V. Guizilini, R. Ambrus, S. Pillai, and A. Gaidon
[ pdf, video, ICML'19 SSL wkshp ]
As there are many differences compared to original paper and the author released their training and testing code, I deleted code at master branch but keep going at dev branch, you can go to packnet-sfm(code published) for all detail.
Hi, Thank you for code!
I could not see reconstruction code. Would you say&share steps for reconstruction (as we can see in the last part of the video)?
Thanks in advance
| Abs Rel | Sq Rel | RMSE | RMSE(log) | Acc.1 | Acc.2 | Acc.3 |
|---|---|---|---|---|---|---|
| 0.119 | 0.890 | 4.855 | 0.198 | 0.862 | 0.954 | 0.980 |
| 这个精度基本相当于monodepth2 的水平。 | ||||||
| 是不是因为为了适应 RTX2060 8G，降低模型训练参数导致的? | ||||||
| 需要设备支持吗？很期待更好的训练预测结果。 |
Hi, thanks for sharing your implementation!
I was just wondering, why is the velocity loss (in trainer.py) commented out? Is it that it doesn't seem to work as well as the paper claims yet?
How do I train with my own dataset? Can I complete my training directly using the continuous monocular video I collected? I don't have radar, speed, etc., what specific training data does he need?
Hello:
Thanks for your excellent work!
But I am curious how much frame rate the model can reach after reducing the amount of parameters?
I need a monodepth estimation to achieve at least 88% accuracy from 80 - 100m , may it possible for now ?
In PackResNetEncoder x = (input_image - 0.45) / 0.225 image norm? why i don't see it in the paper? and more details I cannot find it from the paper.
你好，非常感谢你的分享
请问有关相机内参学习的相关代码具体在哪部分？
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 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.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
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.
