#!/usr/bin/env python3 # # Copyright (c) 2022, NVIDIA CORPORATION. All rights reserved. # # Permission is hereby granted, free of charge, to any person obtaining a # copy of this software and associated documentation files (the "Software"), # to deal in the Software without restriction, including without limitation # the rights to use, copy, modify, merge, publish, distribute, sublicense, # and/or sell copies of the Software, and to permit persons to whom the # Software is furnished to do so, subject to the following conditions: # # The above copyright notice and this permission notice shall be included in # all copies or substantial portions of the Software. # # THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR # IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, # FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL # THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER # LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING # FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER # DEALINGS IN THE SOFTWARE. # import sys import argparse from jetson_inference import backgroundNet from jetson_utils import (videoSource, videoOutput, loadImage, Log, cudaAllocMapped, cudaMemcpy, cudaResize, cudaOverlay) # parse the command line parser = argparse.ArgumentParser(description="Perform background subtraction/removal and replacement.", formatter_class=argparse.RawTextHelpFormatter, epilog=backgroundNet.Usage() + videoSource.Usage() + videoOutput.Usage() + Log.Usage()) parser.add_argument("input_URI", type=str, default="", nargs='?', help="URI of the input stream") parser.add_argument("output_URI", type=str, default="", nargs='?', help="URI of the output stream") parser.add_argument("--network", type=str, default="u2net", help="pre-trained model to load (see below for options)") parser.add_argument("--replace", type=str, default="", help="image filename to use for background replacement") parser.add_argument("--filter-mode", type=str, default="linear", choices=["point", "linear"], help="filtering mode used during visualization, options are:\n 'point' or 'linear' (default: 'linear')") try: args = parser.parse_known_args()[0] except: print("") parser.print_help() sys.exit(0) # load the background removal network net = backgroundNet(args.network, sys.argv) # create video sources & outputs input = videoSource(args.input_URI, argv=sys.argv) output = videoOutput(args.output_URI, argv=sys.argv) # image replacement routines if args.replace: img_replacement = loadImage(args.replace, format='rgba8') img_replacement_scaled = None img_output = None def replaceBackground(img_input): global img_replacement_scaled global img_output if not img_replacement_scaled or img_input.shape != img_replacement_scaled.shape: img_replacement_scaled = cudaAllocMapped(like=img_input) img_output = cudaAllocMapped(like=img_input) cudaResize(img_replacement, img_replacement_scaled, filter=args.filter_mode) cudaMemcpy(img_output, img_replacement_scaled) cudaOverlay(img_input, img_output, 0, 0) return img_output # process frames until EOS or the user exits while True: # capture the next image (with alpha channel) img_input = input.Capture(format='rgba8') if img_input is None: # timeout continue # perform background removal net.Process(img_input, filter=args.filter_mode) # perform background replacement if args.replace: img_output = replaceBackground(img_input) else: img_output = img_input # render the image output.Render(img_output) # update the title bar output.SetStatus("backgroundNet {:s} | Network {:.0f} FPS".format(net.GetNetworkName(), net.GetNetworkFPS())) # print out performance info net.PrintProfilerTimes() # exit on input/output EOS if not input.IsStreaming() or not output.IsStreaming(): break