tencentarc / hosnerf Goto Github PK
View Code? Open in Web Editor NEWHOSNeRF: Dynamic Human-Object-Scene Neural Radiance Fields from a Single Video
License: Apache License 2.0
HOSNeRF: Dynamic Human-Object-Scene Neural Radiance Fields from a Single Video
License: Apache License 2.0
Thank you for releasing the code!
I'd like to train HOSNerf on the released dataset but I found no description about flow file format.
What is the correct flow file format & file name generated by RAFT?
Hello, I'm really interested in your research.
I want to train my custom video. But, I can't find the preprocessing code.
Could you offer the preprocessing code?
Thanks a lot
Hi, will you release the code of HOSNeRF ๏ผ
Hi, thank you for releasing the code!! This is an exciting project.
I'm wondering if there's an inference script which makes it easy to test this method? E.g. a script to process a video and output an artifact, and then a second script to use the artifact + camera position to output an image?
Also, apologies if this is documented somewhere, but roughly how long does the video processing take on an A100 or similar? And also rendering an image based on the processed video artifact?
Hi, thank you for the nice work.
I wonder if you can provide the trained checkpoints for the 6 sequences so we can run inference directly?
Hi,
Thanks for sharing your great paper and code.
I wish running your pre-trained checkpoint from: https://drive.google.com/drive/folders/15I7z7qjBL6rQ3z91_rzhR284L1vfxxOX
which step does this checkpoint belongs to? is it step 3 (3rd_Complete_HOSNeRF)?
can you please provide an example script for rendering a scene (preferbly with human+background) using your checkpoint?
I tried to modify the code of 3rd_Complete_HOSNeRF/run.py to load Backpack.cpkt checkpoint (not sure what I do is correct) but I am getting below error:
self._restore_modules_and_callbacks(ckpt_path)
File "/home/ofer/miniconda3/lib/python3.11/site-packages/pytorch_lightning/trainer/trainer.py", line 1142, in _restore_modules_and_callbacks
self._checkpoint_connector.restore_model()
File "/home/ofer/miniconda3/lib/python3.11/site-packages/pytorch_lightning/trainer/connectors/checkpoint_connector.py", line 179, in restore_model
self.trainer.strategy.load_model_state_dict(self._loaded_checkpoint)
File "/home/ofer/miniconda3/lib/python3.11/site-packages/pytorch_lightning/strategies/strategy.py", line 319, in load_model_state_dict
self.lightning_module.load_state_dict(checkpoint["state_dict"])
File "/home/ofer/miniconda3/lib/python3.11/site-packages/torch/nn/modules/module.py", line 2153, in load_state_dict
raise RuntimeError('Error(s) in loading state_dict for {}:\n\t{}'.format(
RuntimeError: Error(s) in loading state_dict for LitMipNeRF360:
Unexpected key(s) in state_dict: "human.human_stateembeds.1", "human.human_stateembeds.2".
In call to configurable 'run' (<function run at 0x7f7f934b5bc0>)
can you please assist?
thanks alot,
Ofer
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. ๐๐๐
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.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.