| Name | quicktake JSON |
| Version |
0.0.16
JSON |
| download |
| home_page | |
| Summary | Off-the-shelf computer vision ML models. Yolov5, gender and age determination. |
| upload_time | 2023-08-25 12:14:03 |
| maintainer | |
| docs_url | None |
| author | Zach Wolpe |
| requires_python | >=3.8 |
| license | |
| keywords |
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
# QuickTake
Off-the-shelf computer vision ML models. Yolov5, gender and age determination.
The goal of this repository is to provide, easy to use, abstracted, APIs to powerful computer vision models.
## Models
$3$ models are currently available:
- `Object detection`
- `Gender determination`
- `Age determination`
## Model Engine
The models
- `YoloV5`: Object detection. This forms the basis of the other models. Pretrained on `COCO`. Documentation [here](https://pjreddie.com/darknet/yolo/)
- `Gender`: `ResNet18` is used as the models backend. Transfer learning is applied to model gender. The additional gender training was done on the [gender classification dataset](https://www.kaggle.com/datasets/cashutosh/gender-classification-dataset), using code extract from [here](https://github.com/ndb796/Face-Gender-Classification-PyTorch/blob/main/Face_Gender_Classification_using_Transfer_Learning_with_ResNet18.ipynb).
- `Age`: The age model is an implementation of the `SSR-Net` paper: [SSR-Net: A Compact Soft Stagewise Regression Network for Age Estimation](https://www.ijcai.org/proceedings/2018/0150.pdf). The `pyTorch` model was largely derived from [oukohou](https://github.com/oukohou/SSR_Net_Pytorch/blob/master/inference_images.py).
## Getting Started
Install the package with pip:
````pip install quicktake````
## Usage
Build an instance of the class:
```python
from quicktake import QuickTake
```
#### Image Input
Each model is designed to handle $3$ types of input:
- `raw pixels (torch.Tensor)`: raw pixels of a single image. Used when streaming video input.
- `image path (str)`: path to an image. Used when processing a single image.
- `image directory (str)`: path to a directory of images. Used when processing a directory of images.
### Expected Use
`Gender` and `age` determination models are trained on faces. They work fine on a larger image, however, will fail to make multiple predictions in the case of multiple faces in a single image.
The API is currently designed to chain models:
1. `yolo` is used to identify objects.
2. `IF` a person is detected, the `gender` and `age` models are used to make predictions.
This is neatly bundled in the `QuickTake.yolo_loop()` method.
#### Getting Started
Launch a webcam stream:
```python
QL = QuickTake()
QL.launchStream()
```
_*Note*_: Each model returns the results `results_` as well as the runtime `time_`.
Run on a single frame:
```python
from IPython.display import display
from PIL import Image
import cv2
# example images
img = './data/random/dave.png'
# to avoid distractions
import warnings
warnings.filterwarnings('ignore')
# init module
from quicktake import QuickTake
qt = QuickTake()
# extract frame from raw image path
frame = qt.read_image(img)
```
We can now fit `qt.age(<frame>)` or `qt.gender(<frame>)` on the frame. Alternatively we can cycle through the objects detected by `yolo` and if a person is detected, fit `qt.age()` and `qt.gender()`:
```python
# generate points
for _label, x0,y0,x1,y1, colour, thickness, results, res_df, age_, gender_ in qt.yolo_loop(frame):
_label = QuickTake.generate_yolo_label(_label)
QuickTake.add_block_to_image(frame, _label, x0,y0,x1,y1, colour=colour, thickness=thickness)
```
The result is an image with the bounding boxes and labels, confidence (in yolo prediction), age, and gender if a person is detected.
.
The staged output is also useful:
.
For a more comprehensive _example_ directory.
## Future
I have many more models; deployment methods & applications in the pipeline.
If you wish to contribute, please email me _@zachcolinwolpe@gmail.com_.
Raw data
{
"_id": null,
"home_page": "",
"name": "quicktake",
"maintainer": "",
"docs_url": null,
"requires_python": ">=3.8",
"maintainer_email": "",
"keywords": "",
"author": "Zach Wolpe",
"author_email": "zachcolinwolpe@gmail.com",
"download_url": "https://files.pythonhosted.org/packages/e8/30/7261a1626f5f87cd4ba45959e4aebc96d3b317c30f30d7c0089a249df699/quicktake-0.0.16.tar.gz",
"platform": null,
"description": "# QuickTake\n\nOff-the-shelf computer vision ML models. Yolov5, gender and age determination.\n\nThe goal of this repository is to provide, easy to use, abstracted, APIs to powerful computer vision models.\n\n\n## Models\n\n$3$ models are currently available:\n\n- `Object detection`\n- `Gender determination`\n- `Age determination`\n\n## Model Engine\n\nThe models \n- `YoloV5`: Object detection. This forms the basis of the other models. Pretrained on `COCO`. Documentation [here](https://pjreddie.com/darknet/yolo/)\n- `Gender`: `ResNet18` is used as the models backend. Transfer learning is applied to model gender. The additional gender training was done on the [gender classification dataset](https://www.kaggle.com/datasets/cashutosh/gender-classification-dataset), using code extract from [here](https://github.com/ndb796/Face-Gender-Classification-PyTorch/blob/main/Face_Gender_Classification_using_Transfer_Learning_with_ResNet18.ipynb).\n\n- `Age`: The age model is an implementation of the `SSR-Net` paper: [SSR-Net: A Compact Soft Stagewise Regression Network for Age Estimation](https://www.ijcai.org/proceedings/2018/0150.pdf). The `pyTorch` model was largely derived from [oukohou](https://github.com/oukohou/SSR_Net_Pytorch/blob/master/inference_images.py).\n\n## Getting Started\n\nInstall the package with pip:\n\n````pip install quicktake````\n\n\n## Usage\n\nBuild an instance of the class:\n\n```python\nfrom quicktake import QuickTake\n```\n\n#### Image Input\n\nEach model is designed to handle $3$ types of input:\n\n- `raw pixels (torch.Tensor)`: raw pixels of a single image. Used when streaming video input.\n- `image path (str)`: path to an image. Used when processing a single image.\n- `image directory (str)`: path to a directory of images. Used when processing a directory of images.\n\n### Expected Use\n\n`Gender` and `age` determination models are trained on faces. They work fine on a larger image, however, will fail to make multiple predictions in the case of multiple faces in a single image.\n\nThe API is currently designed to chain models:\n\n1. `yolo` is used to identify objects.\n2. `IF` a person is detected, the `gender` and `age` models are used to make predictions.\n\nThis is neatly bundled in the `QuickTake.yolo_loop()` method.\n\n#### Getting Started\n\nLaunch a webcam stream:\n\n```python\nQL = QuickTake()\nQL.launchStream()\n```\n\n_*Note*_: Each model returns the results `results_` as well as the runtime `time_`.\n\nRun on a single frame:\n\n```python\nfrom IPython.display import display\nfrom PIL import Image\nimport cv2\n\n# example images\nimg = './data/random/dave.png'\n\n# to avoid distractions\nimport warnings\nwarnings.filterwarnings('ignore')\n\n# init module\nfrom quicktake import QuickTake\nqt = QuickTake()\n\n# extract frame from raw image path\nframe = qt.read_image(img)\n```\n\nWe can now fit `qt.age(<frame>)` or `qt.gender(<frame>)` on the frame. Alternatively we can cycle through the objects detected by `yolo` and if a person is detected, fit `qt.age()` and `qt.gender()`:\n\n```python\n# generate points\nfor _label, x0,y0,x1,y1, colour, thickness, results, res_df, age_, gender_ in qt.yolo_loop(frame):\n _label = QuickTake.generate_yolo_label(_label)\n QuickTake.add_block_to_image(frame, _label, x0,y0,x1,y1, colour=colour, thickness=thickness)\n```\n\nThe result is an image with the bounding boxes and labels, confidence (in yolo prediction), age, and gender if a person is detected.\n\n.\n\nThe staged output is also useful:\n\n.\n\n\nFor a more comprehensive _example_ directory. \n\n\n## Future\n\nI have many more models; deployment methods & applications in the pipeline.\n\nIf you wish to contribute, please email me _@zachcolinwolpe@gmail.com_.\n",
"bugtrack_url": null,
"license": "",
"summary": "Off-the-shelf computer vision ML models. Yolov5, gender and age determination.",
"version": "0.0.16",
"project_urls": {
"Source code": "https://github.com/ZachWolpe/QuickTake"
},
"split_keywords": [],
"urls": [
{
"comment_text": "",
"digests": {
"blake2b_256": "0b546a806c2f0cdc7a6a51ddaac84eaa171f6746a0ae91a9ab49f6ea32f0fa7a",
"md5": "b5e4cfa2bd7666515848b185e3a8d569",
"sha256": "30e65fdfccf844adacf260e0ea938690bd0f69bd466fdb2fc264330541177dc2"
},
"downloads": -1,
"filename": "quicktake-0.0.16-py3-none-any.whl",
"has_sig": false,
"md5_digest": "b5e4cfa2bd7666515848b185e3a8d569",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.8",
"size": 6445,
"upload_time": "2023-08-25T12:14:01",
"upload_time_iso_8601": "2023-08-25T12:14:01.661709Z",
"url": "https://files.pythonhosted.org/packages/0b/54/6a806c2f0cdc7a6a51ddaac84eaa171f6746a0ae91a9ab49f6ea32f0fa7a/quicktake-0.0.16-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": "",
"digests": {
"blake2b_256": "e8307261a1626f5f87cd4ba45959e4aebc96d3b317c30f30d7c0089a249df699",
"md5": "a18258322f2b296c2e5aaf50001cc5b7",
"sha256": "249891f1c1e7e0a01f285afd59cd5fc918392fd62c42ebd4e62986c634021633"
},
"downloads": -1,
"filename": "quicktake-0.0.16.tar.gz",
"has_sig": false,
"md5_digest": "a18258322f2b296c2e5aaf50001cc5b7",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.8",
"size": 6327,
"upload_time": "2023-08-25T12:14:03",
"upload_time_iso_8601": "2023-08-25T12:14:03.409822Z",
"url": "https://files.pythonhosted.org/packages/e8/30/7261a1626f5f87cd4ba45959e4aebc96d3b317c30f30d7c0089a249df699/quicktake-0.0.16.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2023-08-25 12:14:03",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "ZachWolpe",
"github_project": "QuickTake",
"github_fetch_exception": true,
"lcname": "quicktake"
}
