Comments (6)
Hi, thank you for the kind words! We have released our rendering script here: https://github.com/cvlab-columbia/zero123/blob/main/objaverse-rendering/scripts/render.sh
Also checkout the original objaverse-rendering repo: https://github.com/allenai/objaverse-rendering
And the downloading instruction of objaverse: https://objaverse.allenai.org/docs/download
Hope this helps!
from zero123.
nice to hear the feedback.
Emmm, so this project do really using the whole objaverse
dataset (800k objects * 12views/each -> ~10M images as mentioned in the paper). Wow, I remember the whole dataset is ~8.5T, my disk...
from zero123.
Yes, due to the size of the dataset it's very hard for us to serve it somewhere for download. We'll see what we can do and keep you updated. The renderings are over 1.5T in size so could hopefully fit. Not easy to finetune foundation models but we try our best!
from zero123.
Yeah I agree, ft the foundation model is a tech. Have'u tried use smaller dataset (to check the performance degradation)?
from zero123.
The renderings would be a wonderful resource - maybe HF will host them?
from zero123.
We haven't tried finetuning on smaller dataset than Objaverse but it'll be interesting experiments to try! Currently no plan to host renderings as Objaverse is actively increasing in size. Our copy will most likely be outdated in a few months anyways.
from zero123.
Related Issues (20)
- Zero123 autoregressive generation
- color mesh structure HOT 1
- The dataset file is too large,about 1.5TB, it's hard to download.
- focal length and principal point of the dataset
- Can export the mesh with texture
- getting some cv2 related error HOT 1
- Cant figure out where to put the model
- sd basemodel HOT 1
- UserWarning: ModelCheckpoint(monitor=’val/loss_simple_ema’) not found in the returned metrics HOT 1
- how to create_carvekit_interface without the online access to huggingface?
- Partial dataset HOT 7
- RuntimeError: DataLoader worker (pid(s) 3360242, 3361527, 3362405) exited unexpectedly HOT 4
- test/benchmark on non-object data
- Can we continue training on the basis of 105000.ckpt? HOT 1
- Run time error `python gradio_new.py` HOT 2
- demo is down
- Partial Objaverse dataset for demo HOT 3
- batch size issue when trianing custom dataset HOT 1
- offline rendering or online rendering
- How to change the view of the input picture
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 zero123.