Comments (1)
Well, from a geometry perspective nothing really changes w.r.t resolution. So you can imagine all parameters except for the internal camera matrix everything stays the same. The only things that change in the internal camera matrix are the camera centre and the focal length. But the only place where the camera is used is when we obtain the renderings. So you might as well use everything as is and upsample the 256X256 renderings. Stretching even further, The renderings really do not contain information that is only visible in very high resolution and are not visible at 256X256. So all in all you might as well keep everything the same on the DECA side and train with 1024X1024 images.
from gif.
Related Issues (20)
- training problems HOT 3
- train on my own datasets,the embd_reg_l loss is 0
- output images are fine, but the meshes are all black HOT 5
- Use DECA to get FLAME parameters, but when using these parameters to generate image, the saved meshes still all black
- Mtl file does not exist HOT 2
- done HOT 2
- done
- done
- train on my own datasets ,but the test results are bad HOT 1
- can't don't load inputfile and checkpoint
- Problem with run generate_random_samples.py HOT 3
- trouble with running generate_random_samples.py HOT 1
- [Q]. How to get images of similar appearance but different pose face images like fig.1 in the paper? HOT 1
- run fail HOT 3
- CUDA out of memory HOT 2
- Download the FLAME_texture_data HOT 1
- ModuleNotFoundError: No module named 'my_utils.photometric_optimization.models' HOT 4
- error: verts, faces, aux = load_obj(obj_filename) HOT 3
- File is missing
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 gif.