update local webcam video matting demo and README

README.md

@@ -10,7 +10,7 @@
 </p>
 
 <div align="center">This is the official project of our paper <b>Is a Green Screen Really Necessary for Real-Time Portrait Matting?</b></div>
-<div align="center">MODNet is a <b>trimap-free</b> model for portrait matting in <b>real time</b> (on a single GPU).</div>
+<div align="center">MODNet is a <b>trimap-free</b> model for portrait matting in <b>real time</b> under <b>changing scenes</b>.</div>
 
 
 ---
@@ -23,7 +23,7 @@
 
 ## Video Matting Demo 
 We provide two real-time portrait video matting demos based on WebCam.   
-If you have an Ubuntu system, we recommend you to try the [offline demo](demo/video_matting) to get a higher *fps*. Otherwise, you can access the [online Colab demo](https://colab.research.google.com/drive/1Pt3KDSc2q7WxFvekCnCLD8P0gBEbxm6J?usp=sharing).
+If you have an Ubuntu system, we recommend you to try the [offline demo](demo/video_matting/webcam) to get a higher *fps*. Otherwise, you can access the [online Colab demo](https://colab.research.google.com/drive/1Pt3KDSc2q7WxFvekCnCLD8P0gBEbxm6J?usp=sharing).
 
 
 ## Image Matting Demo

demo/video_matting/webcam/README.md

@@ -1,18 +1,17 @@
 ## MODNet - WebCam-Based Portrait Video Matting Demo
-This is a MODNet portrait video matting demo based on WebCam. It will call your local WebCam and display the matting results in real time.
+This is a MODNet portrait video matting demo based on WebCam. It will call your local WebCam and display the matting results in real time. The demo can run under CPU or GPU.
 
 ### 1. Requirements
 The basic requirements for this demo are:
 - Ubuntu System
 - WebCam
-- Nvidia GPU with CUDA
 - Python 3+
 
 **NOTE**: If your device does not satisfy the above conditions, please try our [online Colab demo](https://colab.research.google.com/drive/1Pt3KDSc2q7WxFvekCnCLD8P0gBEbxm6J?usp=sharing).
 
 
 ### 2. Introduction
-We use ~400 unlabeled video clips (divided into ~50,000 frames) downloaded from the internet to perform SOC to adapt MODNet to the video domain. Nonetheless, due to insufficient labeled training data (~3k labeled foregrounds), our model may still make errors in portrait semantics estimation under challenging scenes. Besides, this demo does not currently support the OFD trick, which will be provided soon.
+We use ~400 unlabeled video clips (divided into ~50,000 frames) downloaded from the internet to perform SOC to adapt MODNet to the video domain. **Nonetheless, due to insufficient labeled training data (~3k labeled foregrounds), our model may still make errors in portrait semantics estimation under challenging scenes.** Besides, this demo does not currently support the OFD trick, which will be provided soon.
 
 For a better experience, please:
 
@@ -33,18 +32,18 @@ We recommend creating a new conda virtual environment to run this demo, as follo
 2. Download the pre-trained model from this [link](https://drive.google.com/file/d/1Nf1ZxeJZJL8Qx9KadcYYyEmmlKhTADxX/view?usp=sharing) and put it into the folder `MODNet/pretrained/`.
 
 
-3. Create a conda virtual environment named `modnet-webcam` and activate it:
+3. Create a conda virtual environment named `modnet` (if it doesn't exist) and activate it:
      ```
-    conda create -n modnet-webcam python=3.6
-    source activate modnet-webcam
+    conda create -n modnet python=3.6
+    source activate modnet
     ```
 
 4. Install the required python dependencies (here we use PyTorch==1.0.0):
     ```
-    pip install -r demo/video_matting/requirements.txt
+    pip install -r demo/video_matting/webcam/requirements.txt
     ```
 
 5. Execute the main code:
     ```
-    python -m demo.video_matting.webcam
+    python -m demo.video_matting.webcam.run
     ```

demo/video_matting/webcam/run.py

@@ -19,8 +19,17 @@ torch_transforms = transforms.Compose(
 print('Load pre-trained MODNet...')
 pretrained_ckpt = './pretrained/modnet_webcam_portrait_matting.ckpt'
 modnet = MODNet(backbone_pretrained=False)
-modnet = nn.DataParallel(modnet).cuda()
-modnet.load_state_dict(torch.load(pretrained_ckpt))
+modnet = nn.DataParallel(modnet)
+
+GPU = True if torch.cuda.device_count() > 0 else False
+if GPU:
+    print('Use GPU...')
+    modnet = modnet.cuda()
+    modnet.load_state_dict(torch.load(pretrained_ckpt))
+else:
+    print('Use CPU...')
+    modnet.load_state_dict(torch.load(pretrained_ckpt, map_location=torch.device('cpu')))
+
 modnet.eval()
 
 print('Init WebCam...')