Add model card for NAS3R

README.md

@@ -0,0 +1,60 @@
+---
+pipeline_tag: image-to-3d
+---
+
+# NAS3R: From None to All: Self-Supervised 3D Reconstruction via Novel View Synthesis
+
+**NAS3R** is a self-supervised feed-forward framework that jointly learns explicit 3D geometry and camera parameters with no ground-truth annotations and no pretrained priors.
+
+- **Paper:** [From None to All: Self-Supervised 3D Reconstruction via Novel View Synthesis](https://huggingface.co/papers/2603.27455)
+- **Project Page:** [https://ranrhuang.github.io/nas3r/](https://ranrhuang.github.io/nas3r/)
+- **GitHub:** [https://github.com/ranrhuang/NAS3R](https://github.com/ranrhuang/NAS3R)
+
+![Teaser](https://github.com/ranrhuang/ranrhuang.github.io/raw/master/nas3r/static/images/teaser.jpg)
+
+## Installation
+
+1. Clone NAS3R:
+```bash
+git clone --recurse-submodules https://github.com/ranrhuang/NAS3R.git
+cd NAS3R
+```
+
+2. Create the environment (example using conda):
+```bash
+conda create -n nas3r python=3.11 -y
+conda activate nas3r
+pip install torch==2.5.1 torchvision==0.20.1 --index-url https://download.pytorch.org/whl/cu121
+pip install -r requirements.txt
+pip install -e submodules/diff-gaussian-rasterization
+```
+
+## Pre-trained Checkpoints
+
+| Model name | Training resolutions | Training data | Training settings |
+|:---:|:---:|:---:|:---:|
+| [re10k_nas3r.ckpt](https://huggingface.co/RanranHuang/NAS3R/resolve/main/re10k_nas3r.ckpt) | 256x256 | re10k | RE10K, 2 views |
+
+## Usage: Evaluation
+
+To perform Novel View Synthesis and Pose Estimation on NAS3R (VGGT-based architecture) using the RealEstate10K dataset:
+
+```bash
+# Assuming the weight is downloaded to ./checkpoints/re10k_nas3r.ckpt
+python -m src.main +experiment=nas3r/random/re10k mode=test wandb.name=re10k \
+    dataset/view_sampler@dataset.re10k.view_sampler=evaluation \
+    dataset.re10k.view_sampler.index_path=assets/evaluation_index_re10k.json \
+    checkpointing.load=./checkpoints/re10k_nas3r.ckpt \
+    test.save_image=false 
+```
+
+## Citation
+
+```bibtex
+@article{huang2026nas3r,
+      title={From None to All: Self-Supervised 3D Reconstruction via Novel View Synthesis},
+      author={Ranran Huang and Weixun Luo and Ye Mao and Krystian Mikolajczyk},
+      journal={arXiv preprint arXiv: 2603.27455},
+      year={2026}
+}
+```