Comments (1)
Hi,
Glad you find it useful!
That will depend on your application and what sensors you have available.
Making a 3D model from using this method and only monocular camera images is something we would like to work on, but we have no implementation so far. Here are some ideas:
- Predict depth using a Neural Network model. Combine this with the camera intrinsic parameters to calculate the end points in 3D. May be noise due to the depth estimation.
- Take multiple images of the room from different angles. Use a SfM-solver (Structure from Motion), for example COLMAP, to get camera poses in 3D. Now you may triangulate the 3D positions of the line segment endpoints given that each point is visible from at least 2 different positions. Now, you don't have proper scale since we only work with projections. You might estimate scale by assuming a fixed ceiling height, using IMU or by once again utilize depth estimations from a Neural Network.
- Use a Line segment SfM and include the estimate line segments in the solver. The solution should give you 3D line segments directly, but without scale, just like the previous idea.
- Take a plane estimation network, for example plane-RCNN (there are more recent work). Work out which plane your line segment belongs to and utilize the plane parameters to project the line into the 3D plane and calculate end point positions.
I hope that made sense, otherwise you are welcome to post some follow up questions!
from srw-net.
Related Issues (8)
- Errors building HOT 5
- Deploying model to replicate fails with "build.sh" file not found
- KeyError: 'Non-existent config key: MODEL.PARSING_HEAD.LAYERS' HOT 3
- Missing annotation files for inference on my own images HOT 2
- Prediction quality HOT 3
- RuntimeError: mat1 and mat2 shapes cannot be multiplied (128x58 and 128x64) HOT 4
- MIT License 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 srw-net.