Top AI Repos — open-source AI, indexed and scored
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
Top AI Repos tracks AI repositories on GitHub and answers two different questions about each one: is it moving right now, and would you bet a product on it.
[CVPR 2023] RoomFormer: Two-level Queries for Single-stage Floorplan Reconstruction
| Date | Stars |
|---|---|
| 2026-07-24 | 314 |
| 2026-07-25 | 314 |
| 2026-07-28 | 314 |
| 2026-07-30 | 314 |
| 2026-08-06 | 314 |
Today
— stars today
This week
— stars this week
This month
— stars this month
Momentum
0.0
growth rate 0.00%/day
<div align="center">
<h2 align="center">Connecting the Dots: Floorplan Reconstruction Using Two-Level Queries</h2>
<h3 align="center">CVPR 2023</h3>
<a href="https://n.ethz.ch/~yuayue/">Yuanwen Yue</a>, <a href="https://theodorakontogianni.github.io/">Theodora Kontogianni</a>, <a href="https://igp.ethz.ch/personen/person-detail.html?persid=143986">Konrad Schindler</a>, <a href="https://francisengelmann.github.io/">Francis Engelmann</a>
ETH Zurich
<!--  -->
<img src="./imgs/teaser.jpg" width=80% height=80%>
</div>
This repository provides code, data and pretrained models for **RoomFormer**, a Transformer model for single-stage floorplan reconstruction.
[[Project Webpage](https://ywyue.github.io/RoomFormer/)] [[Paper](https://arxiv.org/abs/2211.15658)] [[Video](https://www.youtube.com/watch?v=yzYe4yVN1NU)]
<details open="open" style='padding: 10px; border-radius:5px 30px 30px 5px; border-style: solid; border-width: 1px;'>
<summary>Table of Contents</summary>
<ol>
<li>
<a href="#abstract">Abstract</a>
</li>
<li>
<a href="#method">Method</a>
</li>
<li>
<a href="#preparation">Preparation</a>
</li>
<li>
<a href="#evaluation">Evaluation</a>
</li>
<li>
<a href="#training">Training</a>
</li>
<li>
<a href="#semantically-rich-floorplan">Semantically-rich Floorplan</a>
</li>
<li>
<a href="#citation">Citation</a>
</li>
<li>
<a href="#acknowledgment">Acknowledgment</a>
</li>
</ol>
</details>
## Abstract
We address 2D floorplan reconstruction from 3D scans. Existing approaches typically employ heuristically designed multi-stage pipelines. Instead, we formulate floorplan reconstruction as a single-stage structured prediction task: find a variable-size set of polygons, which in turn are variable-length sequences of ordered vertices. To solve it we develop a novel Transformer architecture that generates polygons of multiple rooms in parallel, in a holistic manner without hand-crafted intermediate stages. The model features two-level queries for polygons and corners, and includes polygon matching to make the network end-to-end trainable. Our method achieves a new state-of-the-art for two challenging datasets, Structured3D and SceneCAD, along with significantly faster inference than previous methods. Moreover, it can readily be extended to predict additional information, i.e., semantic room types and architectural elements like doors and windows.
## Method
**Illustration of the RoomFormer model**. Given a top-down-view density map of the input point cloud, (a) the feature backbone extracts multi-scale features, adds positional encodings, and flattens them before passing them into the (b) Transformer encoder. (c) The Transformer decoder takes as input our two-level queries, one level for the room polygons (up to M) and one level for their corners (up to N per room polygon). A feed-forward network (FFN) predicts a class c for each query to accommodate for varying numbers of rooms and corners. During training, the polygon matching guarantees optimal assignment between predicted and groundtruth polygons.
## Preparation
### Environment
* The code has been tested on Linux with python 3.8, torch 1.9.0, and cuda 11.1.
* We recommend an installation through conda:
* Create an environment:
```shell
conda create -n roomformer python=3.8
conda activate roomformer
```
* Install pytorch and other required packages:
```shell
# adjust the cuda version accordingly
pip install torch==1.9.0+cu111 torchvision==0.10.0+cu111 -f https://download.pytorch.org/whl/torch_stable.html
pip install -r requirements.txt
```
* Compile the deformable-attention modules (from [deformable-DETR](https://github.com/fundamentalvision/Deformable-DETR)) and the differentiable rasterization module (from [BoundaryFormer](https://github.com/mlpc-ucsd/BoundaryFormer)):Excerpt of 8,817 characters
Read on GitHubWould you bet a product on this? Bounded 0–100 and slow moving.
matched fp:28f1d348c07410b6, topic:deep-learning, topic:pytorch
matched fp:28f1d348c07410b6, topic:transformer, readme:transformer architecture