# fc-clip
**Repository Path**: ByteDance/fc-clip
## Basic Information
- **Project Name**: fc-clip
- **Description**: This repo contains the code for our paper Convolutions Die Hard: Open-Vocabulary Panoptic Segmentation with Single Frozen Convolutional CLIP
- **Primary Language**: Unknown
- **License**: Apache-2.0
- **Default Branch**: main
- **Homepage**: None
- **GVP Project**: No
## Statistics
- **Stars**: 0
- **Forks**: 0
- **Created**: 2023-08-08
- **Last Updated**: 2026-02-24
## Categories & Tags
**Categories**: Uncategorized
**Tags**: None
## README
# Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP (NeurIPS 2023)
[](https://paperswithcode.com/sota/open-vocabulary-panoptic-segmentation-on?p=convolutions-die-hard-open-vocabulary-1)
[](https://paperswithcode.com/sota/open-vocabulary-semantic-segmentation-on-3?p=convolutions-die-hard-open-vocabulary-1)
[](https://paperswithcode.com/sota/open-vocabulary-semantic-segmentation-on?p=convolutions-die-hard-open-vocabulary-1)
[](https://paperswithcode.com/sota/open-vocabulary-semantic-segmentation-on-9?p=convolutions-die-hard-open-vocabulary-1)
[](https://paperswithcode.com/sota/open-vocabulary-semantic-segmentation-on-2?p=convolutions-die-hard-open-vocabulary-1)
[](https://paperswithcode.com/sota/open-vocabulary-semantic-segmentation-on-7?p=convolutions-die-hard-open-vocabulary-1)
[](https://paperswithcode.com/sota/open-vocabulary-semantic-segmentation-on-5?p=convolutions-die-hard-open-vocabulary-1)
[](https://paperswithcode.com/sota/open-vocabulary-semantic-segmentation-on-1?p=convolutions-die-hard-open-vocabulary-1)
This repo contains the code for our paper [**Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP**](https://arxiv.org/abs/2308.02487)
*FC-CLIP* is an universal model for open-vocabulary image segmentation problems, consisting of a class-agnostic segmenter, in-vocabulary classifier, out-of-vocabulary classifier. With everything built upon a shared single frozen convolutional CLIP model, *FC-CLIP* not only achieves state-of-the-art performance on various open-vocabulary segmentation benchmarks, but also enjoys a much lower training (3.2 days with 8 V100) and testing costs compared to prior arts.
## Installation
See [installation instructions](INSTALL.md).
## Getting Started
See [Preparing Datasets for FC-CLIP](datasets/README.md).
See [Getting Started with FC-CLIP](GETTING_STARTED.md).
We also support FC-CLIP with [HuggingFace 🤗 Demo](https://huggingface.co/spaces/fun-research/FC-CLIP)
## Model Zoo
## Citing FC-CLIP
If you use FC-CLIP in your research, please use the following BibTeX entry.
```BibTeX
@inproceedings{yu2023fcclip,
title={Convolutions Die Hard: Open-Vocabulary Segmentation with Single Frozen Convolutional CLIP},
author={Qihang Yu and Ju He and Xueqing Deng and Xiaohui Shen and Liang-Chieh Chen},
booktitle={NeurIPS},
year={2023}
}
```
## Acknowledgement
[Mask2Former](https://github.com/facebookresearch/Mask2Former)
[ODISE](https://github.com/NVlabs/ODISE)