Upload README.md with huggingface_hub

README.md

@@ -8,18 +8,33 @@ tags:
   - raster2seq
 ---
 
-# Raster2Seq Checkpoints
+<div align="center">
+  <h1>Official PyTorch models of "Raster2Seq: Polygon Sequence Generation for Floorplan Reconstruction" <a href="https://arxiv.org/abs/2602.09016"> (SIGGRAPH'26)</a></h1>
+</div>
+
+<div align="center">
+  <a href="https://hao-pt.github.io/" target="_blank">Hao&nbsp;Phung</a> &emsp; &emsp;
+  <a href="https://hao-pt.github.io/" target="_blank">Hadar&nbsp;Averbuch-Elor</a>
+  <br> <br>
+  Cornell University &emsp;
+  <br> <br>
+  <a href="https://cornell-vailab.github.io/Raster2Seq/">[Page]</a> &emsp;&emsp;
+  <a href="https://arxiv.org/abs/2602.09016">[Paper]</a> &emsp;&emsp;
+  <a href="https://github.com/Cornell-VAILab/Raster2Seq">[Code]</a> &emsp;&emsp;
+  <br> <br>
+</div>
 
 This repository hosts Raster2Seq PyTorch checkpoints for polygon sequence generation from rasterized floorplans. Each checkpoint is stored in its own subfolder so users can download all checkpoints or only the subfolder they need.
 
 ## Available Checkpoints
 
-| Checkpoint key | Dataset | RoomF1 | Subfolder |
-| --- | --- | ---: | --- |
-| `s3d-bw` | Structured3D-B | 99.6 | `s3d-bw/` |
-| `cubicasa5k` | CubiCasa5K | 88.7 | `cubicasa5k/` |
-| `raster2graph` | Raster2Graph | 97.0 | `raster2graph/` |
-| `s3d-density` | Structured3D-DensityMap | 99.1 | `s3d-density/` |
+| Checkpoint key     | Dataset                 | RoomF1 | Subfolder           |
+| ------------------ | ----------------------- | -----: | ------------------- |
+| `s3d-bw`           | Structured3D-B          |   99.6 | `s3d-bw/`           |
+| `cubicasa5k`       | CubiCasa5K              |   88.7 | `cubicasa5k/`       |
+| `raster2graph`     | Raster2Graph            |   97.0 | `raster2graph/`     |
+| `raster2graph-512` | Raster2Graph-512        |   98.1 | `Raster2Graph-512/` |
+| `s3d-density`      | Structured3D-DensityMap |   99.1 | `s3d-density/`      |
 
 ## Download All Checkpoints
 
@@ -36,7 +51,7 @@ from huggingface_hub import snapshot_download
 
 local_repo = snapshot_download(
     repo_id="haopt/Raster2Seq",
-    allow_patterns="cubicasa5k/*",
+    allow_patterns="Raster2Graph-512/*",
 )
 ```
 
@@ -45,7 +60,7 @@ local_repo = snapshot_download(
 With the Raster2Seq codebase, users can load by alias:
 
 ```bash
-python eval.py --checkpoint hf:cubicasa5k ...
+python eval.py --checkpoint hf:raster2graph-512 ...
 ```
 
 or download directly:
@@ -53,9 +68,16 @@ or download directly:
 ```python
 from raster2seq_hub import download_checkpoint
 
-ckpt_path = download_checkpoint("cubicasa5k")
+ckpt_path = download_checkpoint("raster2graph-512")
 ```
 
-The helper downloads the matching subfolder and returns the cached `checkpoint.pth` path.
+**Please CITE** our paper and give us a :star: whenever this repository is used to help produce published results or incorporated into other software.
 
-Hugging Face repository IDs use the form `namespace/repo`. Dataset-specific choices are represented by subfolders, not by repo IDs such as `Raster2Seq/cubicasa5k`.
+```bibtex
+@inproceedings{phung2026raster2seq,
+   title={Raster2Seq: Polygon Sequence Generation for Floorplan Reconstruction},
+   author={Phung, Hao and Averbuch-Elor, Hadar},
+   booktitle={Special Interest Group on Computer Graphics and Interactive Techniques Conference Conference Papers},
+   year= {2026},
+}
+```