Comments (2)
Hi @InFistLee, thanks for your interest and questions!
Creating the test renders basically boils down to just rendering the scene from the test camera. You can simply do that by selecting the test camera to be the active one, and then rendering the scene under Render > Render Animation (in the Topbar). Note that by default the rendered images will be stored into the /tmp/
directory, you might want to change that. Pasting the renders into a test
folder within the created BlenderNeRF dataset folder should do the trick.
As for the the test poses for the COS method, these will be the camera pose for the selected Camera (highlighted in blue in the screenshot). If you want to have a test camera with poses sampled from the same sphere, you can directly select the BlenderNeRF Camera as active camera in the COS method (visible once you pressed the Camera button under Preview). In order to get different poses for training and testing cameras with the COS method, I would suggest first only rendering training frames with a specific seed, and then only test frames with a different seed.
Let me know if anything is still unclear or if you have further questions!
from blendernerf.
Also, I found that when using COS method, the sampled testing camera pose is actually not on the sphere and pointing to the center of the scene. I guess that the testing poses are randomly sampled unlike training poses, which might be a problem in 360 scene.
from blendernerf.
Related Issues (20)
- DroNeRF: new paper on determining ideal capture (render) locations: HOT 1
- The output images didn't get whole scene of a model HOT 2
- I export a data set from blender but camera doesnt match in instant NGP HOT 11
- Question regarding GT HOT 6
- Request for Feature Addition: Simultaneous Rendering of Depth Maps with Optical Images and JSON Integration HOT 1
- How to automatic the process without using blender HOT 3
- how to start, please give more detailed documentations? HOT 5
- Is this tool only generate training image directory? HOT 2
- How to calculate transform_matrix? HOT 3
- Call in a for loop HOT 3
- Scene is shifting? HOT 6
- BlenderNeRF Camera not found HOT 5
- Can I use this tool to generate 3D Gaussian splatting datasets? HOT 7
- The color of the object in the generated image is different. HOT 4
- generating test renderings. HOT 5
- TypeError: empty_fn() missing 1 required positional argument: 'context' HOT 2
- glass ceiling on radius parameter HOT 1
- Failed to reconstruct with SPEED+ dataset HOT 1
- Request: Support creation of 3DGS datasets HOT 6
- Just a question really 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 blendernerf.