https://github.com/breezedeus/coin-clip
Coin-CLIP: fine-tuned with a vast collection of coin images from CLIP using contrastive learning. It enhances feature extraction for coins, boosting image search accuracy. This model merges Visual Transformer (ViT) with CLIP's multimodal learning, optimized for numismatic applications.
https://github.com/breezedeus/coin-clip
clip coin coin-identification coin-recognition coin-retrieval multimodal-learning numismatics
Last synced: 7 months ago
JSON representation
Coin-CLIP: fine-tuned with a vast collection of coin images from CLIP using contrastive learning. It enhances feature extraction for coins, boosting image search accuracy. This model merges Visual Transformer (ViT) with CLIP's multimodal learning, optimized for numismatic applications.
- Host: GitHub
- URL: https://github.com/breezedeus/coin-clip
- Owner: breezedeus
- License: apache-2.0
- Created: 2023-11-26T08:11:30.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2023-12-08T09:25:21.000Z (over 1 year ago)
- Last Synced: 2024-11-28T16:52:12.158Z (7 months ago)
- Topics: clip, coin, coin-identification, coin-recognition, coin-retrieval, multimodal-learning, numismatics
- Language: Python
- Homepage: https://huggingface.co/spaces/breezedeus/USA-Coin-Retrieval
- Size: 50.3 MB
- Stars: 16
- Watchers: 3
- Forks: 3
- Open Issues: 4
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
Β
[](https://pepy.tech/project/coin_clip)
[](./LICENSE)
[](https://badge.fury.io/py/coin_clip)
[](https://github.com/breezedeus/coin-clip)
[](https://github.com/breezedeus/coin-clip)
[](https://twitter.com/breezedeus)
[𧳠Model](https://huggingface.co/breezedeus/coin-clip-vit-base-patch32) |
[ππ» Online Demo](https://huggingface.co/spaces/breezedeus/USA-Coin-Retrieval) |
[π¬ Contact](https://www.breezedeus.com/join-group)
[[δΈζ]](./README_cn.md) | [English]
# Coin-CLIP πͺ : Enhancing Coin Image Retrieval with CLIP
**[Coin-CLIP](https://huggingface.co/breezedeus/coin-clip-vit-base-patch32)** `breezedeus/coin-clip-vit-base-patch32` is built upon
OpenAI's **[CLIP](https://huggingface.co/openai/clip-vit-base-patch32) (ViT-B/32)** model and fine-tuned on
a dataset of more than `340,000` coin images using contrastive learning techniques. This specialized model is designed to significantly improve feature extraction for coin images, leading to more accurate image-based search capabilities. Coin-CLIP combines the power of Visual Transformer (ViT) with CLIP's multimodal learning capabilities, specifically tailored for the numismatic domain.
**Key Features:**
- State-of-the-art coin image retrieval;
- Enhanced feature extraction for numismatic images;
- Seamless integration with CLIP's multimodal learning.
To further simplify the use of the **Coin-CLIP** model, this project provides tools for quickly building a coin image retrieval engine.
# Comparison: Coin-CLIP vs. CLIP
#### Example 1 (Left: Coin-CLIP; Right: CLIP)
#### Example 2 (Left: Coin-CLIP; Right: CLIP)
#### More Examples
more
Example 3 (Left: Coin-CLIP; Right: CLIP)
Example 4 (Left: Coin-CLIP; Right: CLIP)
Example 5 (Left: Coin-CLIP; Right: CLIP)
Example 6 (Left: Coin-CLIP; Right: CLIP)
# Install
```bash
pip install coin_clip
```
# Usage
## Code Examples
### Extract Feature Vectors from Coin Images
```python
from coin_clip import CoinClip
# Automatically download the model from Huggingface
model = CoinClip(model_name='breezedeus/coin-clip-vit-base-patch32')
images = ['examples/10_back.jpg', 'examples/16_back.jpg']
img_feats, success_ids = model.get_image_features(images)
print(img_feats.shape) # --> (2, 512)
```
> β οΈ **Note**:
>
> The above code automatically downloads the [`breezedeus/coin-clip-vit-base-patch32`](https://huggingface.co/breezedeus/coin-clip-vit-base-patch32) model from Huggingface.
If you cannot download automatically, please manually download the model locally, and then initialize `CoinClip` by specifying the local directory of the model through the `model_name` parameter, like `model_name='path/to/coin-clip-vit-base-patch32'`.
## Command line tools
### Building a Vector Retrieval Engine
`coin-clip build-db` can be used to build a vector search engine. It extracts features from all coin images πͺ in a specified directory and builds a ChromaDB vector search engine.
```bash
$ coin-clip build-db -h
Usage: coin-clip build-db [OPTIONS]
Extract vectors from a candidate image set and build a search engine based
on it.
Options:
-m, --model-name TEXT Model Name; either local path or huggingface
model name [default: breezedeus/coin-clip-vit-
base-patch32]
-d, --device TEXT ['cpu', 'cuda']; Either 'cpu' or 'gpu', or
specify a specific GPU like 'cuda:0'. Default is
'cpu'. [default: cpu]
-i, --input-image-dir TEXT Folder with Coin Images to be indexed. [required]
-o, --output-db-dir TEXT Folder where the built search engine is stored.
[default: ./coin_clip_chroma.db]
-h, --help Show this message and exit.
```
For instance,
```bash
$ coin-clip build-db -i examples -o coin_clip_chroma.db
```
### Querying
After building the vector search engine with the above command, you can use the `coin-clip retrieve` command to retrieve the coin images πͺ most similar to a specified coin image.
```bash
$ coin-clip retrieve -h
Usage: coin-clip retrieve [OPTIONS]
Retrieve images from the search engine, based on the query image.
Options:
-m, --model-name TEXT Model Name; either local path or huggingface model
name [default: breezedeus/coin-clip-vit-base-
patch32]
-d, --device TEXT ['cpu', 'cuda']; Either 'cpu' or 'gpu', or specify a
specific GPU like 'cuda:0'. Default is 'cpu'.
[default: cpu]
--db-dir TEXT Folder where the built search engine is stored.
[default: ./coin_clip_chroma.db]
-i, --image-fp TEXT Image Path to retrieve [required]
-h, --help Show this message and exit.
```
For instance,
```bash
$ coin-clip retrieve --db-dir coin_clip_chroma.db -i examples/10_back.jpg
```
## A cup of coffee for the author
It is not easy to maintain and evolve the project, so if it is helpful to you, please consider [offering the author a cup of coffee π₯€](https://www.breezedeus.com/buy-me-coffee).
---
Official code base: [https://github.com/breezedeus/coin-clip](https://github.com/breezedeus/coin-clip). Please cite it properly.