counting-guidance

KO | EN

gitlite — search

#javascript #python #golang #machine-learning #rust #deep-learning #linux #windows #vue #security #ios #mcp #openai #open-source

counting-guidance

(★ 15)

[WACV 2025 Oral] Counting Guidance for High Fidelity Text-to-Image Synthesis

counting-guidance Latest Version Download

Download Latest Version (.zip)

// repository documentation

Was this content helpful?

★ 0 (0 ratings)

<h1 align="center"> [WACV 2025 Oral] Counting Guidance for High Fidelity Text-to-Image Synthesis</h1>
<h4 align="center">
 
 <a href="https://wonjunn.github.io/" target="_blank">Wonjun Kang</a>1,2*, <a href="https://scholar.google.com/citations?user=G1EpeWYAAAAJ&hl=en" target="_blank">Kevin Galim</a>2*, <a href="http://cvml.ajou.ac.kr/wiki/index.php/Professor" target="_blank">Hyung Il Koo</a>2,3, <a href="https://ece.snu.ac.kr/en/research-faculty/faculty/fulltime?md=view&profid=p041" target="_blank">Nam Ik Cho</a>1
 
 
 1<a href="https://en.snu.ac.kr/index.html" target="_blank">Seoul National University</a>, 2<a href="https://furiosa.ai/" target="_blank">FuriosaAI</a>, 3<a href="https://www.ajou.ac.kr/en/index.do" target="_blank">Ajou University</a>
 
 </h4>

**Paper Link**: https://arxiv.org/abs/2306.17567

![teaser](https://raw.githubusercontent.com/furiosa-ai/counting-guidance/main/docs/images/teaser.png)

## Updates

* [01/20/25] Our paper is selected for oral presentation at <a href="https://wacv2025.thecvf.com/">WACV 2025</a>! 🎉🎉
* [10/28/24] Our paper is accepted by <a href="https://wacv2025.thecvf.com/">WACV 2025</a>! 🎉

## Usage

- Note, we tested the code on a NVIDIA V100 32GB GPU. On different GPUs, results might slightly differ.

### Setup

- Download RCC counting network weights [counting.ckpt](https://drive.google.com/drive/folders/12Ak9uIyLHkRo59zCbEZqXrRe0kP5iz_S?usp=drive_link) to weights/counting.ckpt
- Run
    ```
    conda env create --name ldm_count --file=env.yml
    pip install timm==0.9.2
    conda activate ldm_count
    wget https://huggingface.co/ShilongLiu/GroundingDINO/resolve/main/groundingdino_swint_ogc.pth -P weights
    ```

### Generate one image with counting guidance
```
python generate.py \
    --prompt "six apples on the table" \
    --counting_words apples \
    --counting_word_counts 6 \
    --counting_scale 1.0 \
    --seed 0 \
    --output output.png
```

### Reproduce paper results
```
# Check exp/ for outputs
sh run_metrics.sh
```

## Citation
```tex
@inproceedings{kang2025counting,
  title={Counting guidance for high fidelity text-to-image synthesis},
  author={Kang, Wonjun and Galim, Kevin and Koo, Hyung Il and Cho, Nam Ik},
  booktitle={2025 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)},
  pages={899--908},
  year={2025},
  organization={IEEE}
}
```

counting-guidance

counting-guidance Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

counting-guidance

counting-guidance Latest Version Download

Today's picks

Was this content helpful?

Recent Feedback

Download README