BackgroundMattingV2/inference_webcam_ts_compositing.py

"""
Inference on webcams: Use a model on webcam input.

Once launched, the script is in background collection mode.
Press B to toggle between background capture mode and matting mode. The frame shown when B is pressed is used as background for matting.
Press Q to exit.

Example:

    python inference_webcam.py \
        --model-type mattingrefine \
        --model-backbone resnet50 \
        --model-checkpoint "PATH_TO_CHECKPOINT" \
        --resolution 1280 720

"""

import argparse, os, shutil, time
import cv2
import numpy as np
import torch

from PyQt5 import QtGui, QtCore, uic
from PyQt5 import QtWidgets
from PyQt5.QtWidgets import QMainWindow, QApplication

from torch import nn
from torch.utils.data import DataLoader
from torchvision.transforms import Compose, ToTensor, Resize
from torchvision.transforms.functional import to_pil_image
from threading import Thread, Lock, Condition
from tqdm import tqdm
from PIL import Image, ImageTk

from dataset import VideoDataset
from model import MattingBase, MattingRefine
from inference_webcam import Displayer, FPSTracker, Camera


# --------------- Arguments ---------------


parser = argparse.ArgumentParser(description='Inference from web-cam')

parser.add_argument('--model-type', type=str, required=True, choices=['mattingbase', 'mattingrefine'])
parser.add_argument('--model-backbone', type=str, required=True, choices=['resnet101', 'resnet50', 'mobilenetv2'])
parser.add_argument('--model-backbone-scale', type=float, default=0.25)
parser.add_argument('--model-checkpoint', type=str, required=True)
parser.add_argument('--model-refine-mode', type=str, default='sampling', choices=['full', 'sampling', 'thresholding'])
parser.add_argument('--model-refine-sample-pixels', type=int, default=80_000)
parser.add_argument('--model-refine-threshold', type=float, default=0.7)

parser.add_argument('--hide-fps', action='store_true')
parser.add_argument('--resolution', type=int, nargs=2, metavar=('width', 'height'), default=(1280, 720))

parser.add_argument('--device-id', type=int, default=0)
parser.add_argument('--background-image', type=str, default="")
parser.add_argument('--fps-limit', type=int, default=1000)

args = parser.parse_args()


# --------------- Main ---------------

default_float_dtype = torch.get_default_dtype()
model = torch.jit.load(args.model_checkpoint)
model.backbone_scale = args.model_backbone_scale
model.refine_mode = args.model_refine_mode
model.refine_sample_pixels = args.model_refine_sample_pixels
model.refine_threshold = args.model_refine_threshold

model = model.to(torch.device('cuda'))

width, height = args.resolution
cam = Camera(device_id=args.device_id,width=width, height=height)
app = QApplication(['MattingV2'])
dsp = Displayer('MattingV2', cam.width, cam.height, show_info=(not args.hide_fps))
dsp.show()


def cv2_frame_to_cuda(frame):
    frame = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
    if 'fp16'in args.model_checkpoint:
      return ToTensor()(Image.fromarray(frame)).unsqueeze_(0).to(torch.float16).cuda()
    else:
      pic = Image.fromarray(frame)
      img = torch.ByteTensor(torch.ByteStorage.from_buffer(pic.tobytes()))
      img = img.cuda()
      img = img.view(pic.size[1], pic.size[0], len(pic.getbands()))
      img = img.permute((2, 0, 1)).contiguous()
      tmp = img.to(dtype=default_float_dtype).div(255)
      tmp.unsqueeze_(0)
      tmp = tmp.to(torch.float32)
      return tmp
      #return ToTensor()(Image.fromarray(frame)).unsqueeze_(0).to(torch.float32).cuda()

with torch.no_grad():
    while True:
        bgr = None
        while True: # grab bgr
            frame = cam.read()
            frameRGB = cv2.cvtColor(frame, cv2.COLOR_BGR2RGB)
            key = dsp.step(frameRGB)
            if key == 'b':
              bgr = cv2_frame_to_cuda(frame)
              break
            elif key == 'w':
              cam.brighter()
            elif key == 's':
              cam.darker()
            elif key == 'q':
                exit()

        if args.background_image == "":
          #green screen
          bgImage = torch.zeros_like(bgr)
          bgImage[0,1] = torch.ones_like(bgr[0,0])
        else:
          bgImage = cv2.imread(args.background_image, cv2.IMREAD_UNCHANGED)
          bgImage = cv2.resize(bgImage, (frame.shape[1], frame.shape[0]))
          bgImage = cv2_frame_to_cuda(bgImage)

        while True: # matting
            frame = cam.read()
            src = cv2_frame_to_cuda(frame)
            pha, fgr = model(src, bgr)[:2]
            res = pha * fgr + (1 - pha) * bgImage
            res = res.mul(255).byte().cpu().permute(0, 2, 3, 1).numpy()[0]
            key = dsp.step(res.copy())
            if key == 'b':
                break
            elif key == 'w':
              cam.brighter()
            elif key == 's':
              cam.darker()
            elif key == 'q':
                exit()