Zero-1-to-3: Zero-shot One Image to 3D Object

Project Page | Paper | Weights | Demo (precomputed) | Demo (live)

Zero-1-to-3: Zero-shot One Image to 3D Object
Ruoshi Liu¹, Rundi Wu¹, Basile Van Hoorick¹, Pavel Tokmakov², Sergey Zakharov², Carl Vondrick¹
¹Columbia University, ²Toyota Research Institute

Updates

If you haven't started downloading the dataset, please hold it off for now as the link will be removed shortly due to unexpected high traffic. We will update with new methods of downloading asap.

We have released training script and objaverse rendering

$\text{\color{red}{Live demo released}}$ 🤗: https://huggingface.co/spaces/cvlab/zero123-live. Shout out to Huggingface for funding this demo!!

We've optimized our code base with some simple tricks and the current demo runs at around 22GB VRAM so it's runnable on a RTX 3090/4090(Ti)!

Usage

Novel View Synthesis

conda create -n zero123 python=3.9
conda activate zero123
cd zero123
pip install -r requirements.txt
git clone https://github.com/CompVis/taming-transformers.git
pip install -e taming-transformers/
git clone https://github.com/openai/CLIP.git
pip install -e CLIP/

Download checkpoint under zero123 through one of the following sources:

https://drive.google.com/drive/folders/1geG1IO15nWffJXsmQ_6VLih7ryNivzVs?usp=sharing
https://huggingface.co/cvlab/zero123-weights
wget https://cv.cs.columbia.edu/zero123/assets/$iteration.ckpt    # iteration = [105000, 165000, 230000, 300000]

Note that we have released 4 model weights: 105000.ckpt, 165000.ckpt, 230000.ckpt, 300000.ckpt. By default, we use 105000.ckpt which is the checkpoint after finetuning 105000 iterations on objaverse. Naturally, checkpoints trained longer tend to overfit to training data and suffer in zero-shot generalization, though we didn't empirically verify this. 300000.ckpt is trained for around 6000 A100 hours.

Run our gradio demo for novel view synthesis:

python gradio_new.py

Note that this app uses around 22 GB of VRAM, so it may not be possible to run it on any GPU.

Training Script (preliminary)

Download image-conditioned stable diffusion checkpoint released by Lambda Labs:
wget https://cv.cs.columbia.edu/zero123/assets/sd-image-conditioned-v2.ckpt

Run training command:

python main.py \
    -t \
    --base configs/stable-diffusion/sd-objaverse-finetune-c_concat-256.yaml \
    --gpus 0,1,2,3,4,5,6,7 \
    --scale_lr False \
    --num_nodes 1 \
    --seed 42 \
    --check_val_every_n_epoch 10 \
    --finetune_from sd-image-conditioned-v2.ckpt

Note that this training script is set for an 8-GPU system, each with 80GB of VRAM. As discussed in the paper, empirically the large batch size is very important for "stably" training stable diffusion. If you have smaller GPUs, consider using smaller batch size and gradient accumulation to obtain a similar effective batch size. Please check this thread for the train/val split we used in the paper.

Objaverse Rendering (preliminary)

Download our objaverse renderings with:

wget https://cv.cs.columbia.edu/zero123/assets/views_release.tar.gz

3D Reconstruction

Note that we haven't extensively tuned the hyperparameters for 3D recosntruction. Feel free to explore and play around!

cd 3drec
pip install -r requirements.txt
python run_zero123.py \
    --scene pikachu \
    --index 0 \
    --n_steps 10000 \
    --lr 0.05 \
    --sd.scale 100.0 \
    --emptiness_weight 0 \
    --depth_smooth_weight 10000. \
    --near_view_weight 10000. \
    --train_view True \
    --prefix "experiments/exp_wild" \
    --vox.blend_bg_texture False \
    --nerf_path "data/nerf_wild"

We tested the installation processes on a system with Ubuntu 20.04 with an NVIDIA GPU with Ampere architecture.

Acknowledgement

This repository is based on Stable Diffusion, Objaverse, and SJC. We would like to thank the authors of these work for publicly releasing their code. We would like to thank the authors of NeRDi and SJC for their helpful feedback.

We would like to thank Changxi Zheng and Chengzhi Mao for many helpful discussions. This research is based on work partially supported by the Toyota Research Institute, the DARPA MCS program under Federal Agreement No. N660011924032, and the NSF NRI Award #1925157.

Citation

@misc{liu2023zero1to3,
      title={Zero-1-to-3: Zero-shot One Image to 3D Object}, 
      author={Ruoshi Liu and Rundi Wu and Basile Van Hoorick and Pavel Tokmakov and Sergey Zakharov and Carl Vondrick},
      year={2023},
      eprint={2303.11328},
      archivePrefix={arXiv},
      primaryClass={cs.CV}
}

ruoshiliu/zero123