- Mating is the technique of extracting foreground from an image by calculating its color and transparency. It is widely used in the film industry to replace background, image composition, and visual effects. Each pixel in the image will have a value that represents its foreground transparency, called Alpha. The set of all Alpha values in an image is called Alpha Matte. The part of the image covered by the mask can be extracted to complete foreground separation.
- For more information, please refer to: [dim_vgg16_matting](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.3/contrib/Matting)
## II. Installation
...
...
@@ -46,11 +46,11 @@
- ```shell
$ hub install dim_vgg16_matting
```
- In case of any problems during installation, please refer to:[Windows_Quickstart](../../../../docs/docs_en/get_start/windows_quickstart.md)
$ hub run dim_vgg16_matting --input_path "/PATH/TO/IMAGE" --trimap_path "/PATH/TO/TRIMAP"
```
- If you want to call the Hub module through the command line, please refer to: [PaddleHub Command Line Instruction](../../../../docs/docs_en/tutorial/cmd_usage.rst)
...
...
@@ -70,16 +70,16 @@
model = hub.Module(name="dim_vgg16_matting")
result = model.predict(image_list=["/PATH/TO/IMAGE"], trimap_list=["PATH/TO/TRIMAP"])
result = model.predict(image_list=["/PATH/TO/IMAGE"], trimap_list=["/PATH/TO/IMAGE"])
print(result)
```
- ### 3、API
- ```python
def predict(self,
image_list,
trimap_list,
visualization,
def predict(self,
image_list,
trimap_list,
visualization,
save_path):
```
...
...
@@ -88,7 +88,7 @@
- **Parameter**
- image_list (list(str | numpy.ndarray)): Image path or image data, ndarray.shape is in the format \[H, W, C\]，BGR.
- trimap_list(list(str | numpy.ndarray)): Trimap path or trimap data, ndarray.shape is in the format \[H, W]，Gray style.
- trimap_list(list(str | numpy.ndarray)): Trimap path or trimap data, ndarray.shape is in the format \[H, W]，Gray style.
- visualization (bool): Whether to save the recognition results as picture files, default is False.
- save_path (str): Save path of images, "dim_vgg16_matting_output" by default.
...
...
@@ -96,7 +96,7 @@
- result (list(numpy.ndarray))：The list of model results.
## IV. Server Deployment
- PaddleHub Serving can deploy an online service of matting.
- Mating is the technique of extracting foreground from an image by calculating its color and transparency. It is widely used in the film industry to replace background, image composition, and visual effects. Each pixel in the image will have a value that represents its foreground transparency, called Alpha. The set of all Alpha values in an image is called Alpha Matte. The part of the image covered by the mask can be extracted to complete foreground separation.
- For more information, please refer to: [gfm_resnet34_matting](https://github.com/JizhiziLi/GFM)
## II. Installation
...
...
@@ -46,11 +46,11 @@
- ```shell
$ hub install gfm_resnet34_matting
```
- In case of any problems during installation, please refer to:[Windows_Quickstart](../../../../docs/docs_en/get_start/windows_quickstart.md)
$ hub run gfm_resnet34_matting --input_path "/PATH/TO/IMAGE"
```
- If you want to call the Hub module through the command line, please refer to: [PaddleHub Command Line Instruction](../../../../docs/docs_en/tutorial/cmd_usage.rst)
...
...
@@ -69,16 +69,16 @@
import cv2
model = hub.Module(name="gfm_resnet34_matting")
result = model.predict(["/PATH/TO/IMAGE"])
result = model.predict([cv2.imread("/PATH/TO/IMAGE")])
print(result)
```
- ### 3、API
- ```python
def predict(self,
image_list,
visualization,
def predict(self,
image_list,
visualization,
save_path):
```
...
...
@@ -94,7 +94,7 @@
- result (list(numpy.ndarray))：The list of model results.
## IV. Server Deployment
- PaddleHub Serving can deploy an online service of matting.
