Image Resampling

Overview

The Rescale application rescales the input image by first applying a low-pass filter to avoid aliasing, then doing a downsampling. The resulting image has half of input's width and one third of input's height. The result is then saved to disk.

Instructions

The command line parameters are:

<backend> <input image>

where

• backend: either cpu, cuda or pva; it defines the backend that will perform the processing.
• input image: input image file name to be downsampled, it accepts png, jpeg and possibly others.

Here's one example:

• C++

  ./vpi_sample_04_rescale cuda ../assets/kodim8.png

• Python

  python main.py cuda ../assets/kodim8.png

This is using the CUDA backend and one of the provided sample images. You can try with other images, respecting the constraints imposed by the algorithm.

Results

Input image	Output image, downsampled

Source Code

For convenience, here's the code that is also installed in the samples directory.

Language: C++ Python

import sys
import vpi
import numpy as np
from PIL import Image
from argparse import ArgumentParser
 
# Parse command line arguments
parser = ArgumentParser()
parser.add_argument('backend', choices=['cpu','cuda','vic'],
                    help='Backend to be used for processing')
 
parser.add_argument('input',
                    help='Image to be used as input')
 
args = parser.parse_args();
 
if args.backend == 'cpu':
    backend = vpi.Backend.CPU
elif args.backend == 'cuda':
    backend = vpi.Backend.CUDA
else:
    assert args.backend == 'vic'
    backend = vpi.Backend.VIC
 
# Load input into a vpi.Image
input = vpi.asimage(np.asarray(Image.open(args.input)))
 
# Using the chosen backend,
with backend:
    # First convert input to NV12_ER.
    # We're overriding the default backend with CUDA.
    temp = input.convert(vpi.Format.NV12_ER, backend=vpi.Backend.CUDA)
 
    # Rescale the image using the chosen backend
    temp = temp.rescale((input.width//2, input.height//3))
 
    # Convert result back to input's format
    output = temp.convert(input.format, backend=vpi.Backend.CUDA)
 
# Save result to disk
Image.fromarray(output.cpu()).save('scaled_python'+str(sys.version_info[0])+'_'+args.backend+'.png')
 
# vim: ts=8:sw=4:sts=4:et:ai

#include <opencv2/core/version.hpp>
#if CV_MAJOR_VERSION >= 3
#    include <opencv2/imgcodecs.hpp>
#else
#    include <opencv2/highgui/highgui.hpp>
#endif
 
#include <vpi/OpenCVInterop.hpp>
 
#include <vpi/Image.h>
#include <vpi/Status.h>
#include <vpi/Stream.h>
#include <vpi/algo/ConvertImageFormat.h>
#include <vpi/algo/Rescale.h>
 
#include <cstring> // for memset
#include <iostream>
#include <sstream>
 
#define CHECK_STATUS(STMT)                                    \
    do                                                        \
    {                                                         \
        VPIStatus status = (STMT);                            \
        if (status != VPI_SUCCESS)                            \
        {                                                     \
            char buffer[VPI_MAX_STATUS_MESSAGE_LENGTH];       \
            vpiGetLastStatusMessage(buffer, sizeof(buffer));  \
            std::ostringstream ss;                            \
            ss << vpiStatusGetName(status) << ": " << buffer; \
            throw std::runtime_error(ss.str());               \
        }                                                     \
    } while (0);
 
int main(int argc, char *argv[])
{
    // OpenCV image that will be wrapped by a VPIImage.
    // Define it here so that it's destroyed *after* wrapper is destroyed
    cv::Mat cvImage;
 
    // VPI objects that will be used
    VPIImage image      = NULL;
    VPIImage imageNV12  = NULL;
    VPIImage outputNV12 = NULL;
    VPIImage output     = NULL;
    VPIStream stream    = NULL;
 
    int retval = 0;
 
    try
    {
        if (argc != 3)
        {
            throw std::runtime_error(std::string("Usage: ") + argv[0] + " <cpu|vic|cuda> <input image>");
        }
 
        std::string strBackend       = argv[1];
        std::string strInputFileName = argv[2];
 
        // Load the input image
        cvImage = cv::imread(strInputFileName);
        if (cvImage.empty())
        {
            throw std::runtime_error("Can't open '" + strInputFileName + "'");
        }
 
        assert(cvImage.type() == CV_8UC3);
 
        // Now parse the backend
        VPIBackend backend;
 
        if (strBackend == "cpu")
        {
            backend = VPI_BACKEND_CPU;
        }
        else if (strBackend == "cuda")
        {
            backend = VPI_BACKEND_CUDA;
        }
        else if (strBackend == "vic")
        {
            backend = VPI_BACKEND_VIC;
        }
        else
        {
            throw std::runtime_error("Backend '" + strBackend + "' not recognized, it must be either cpu, cuda or vic");
        }
 
        // 1. Initialization phase ---------------------------------------
 
        // Create the stream for the given backend. We'll also enable CUDA for gaussian filter.
        CHECK_STATUS(vpiStreamCreate(backend | VPI_BACKEND_CUDA, &stream));
 
        // We now wrap the loaded image into a VPIImage object to be used by VPI.
        // VPI won't make a copy of it, so the original
        // image must be in scope at all times.
        CHECK_STATUS(vpiImageCreateOpenCVMatWrapper(cvImage, 0, &image));
 
        // Create a temporary image to hold the input converted to NV12.
        CHECK_STATUS(vpiImageCreate(cvImage.cols, cvImage.rows, VPI_IMAGE_FORMAT_NV12_ER, 0, &imageNV12));
 
        // Now create the output image.
        CHECK_STATUS(vpiImageCreate(cvImage.cols / 2, cvImage.rows / 3, VPI_IMAGE_FORMAT_NV12_ER, 0, &outputNV12));
 
        // And the output image converted back to BGR8
        CHECK_STATUS(vpiImageCreate(cvImage.cols / 2, cvImage.rows / 3, VPI_IMAGE_FORMAT_BGR8, 0, &output));
 
        // 2. Computation phase ---------------------------------------
 
        // Convert input from BGR8 to NV12
        CHECK_STATUS(vpiSubmitConvertImageFormat(stream, VPI_BACKEND_CUDA, image, imageNV12, NULL));
 
        // Now we downsample
        CHECK_STATUS(vpiSubmitRescale(stream, backend, imageNV12, outputNV12, VPI_INTERP_LINEAR, VPI_BORDER_CLAMP, 0));
 
        // Finally, convert the result back to BGR8
        CHECK_STATUS(vpiSubmitConvertImageFormat(stream, VPI_BACKEND_CUDA, outputNV12, output, NULL));
 
        // Wait until the algorithm finishes processing
        CHECK_STATUS(vpiStreamSync(stream));
 
        // Now let's retrieve the output image contents and output it to disk
        {
            VPIImageData data;
            CHECK_STATUS(vpiImageLock(output, VPI_LOCK_READ, &data));
 
            // Lock output image to retrieve its data on cpu memory
            VPIImageData outData;
            CHECK_STATUS(vpiImageLock(output, VPI_LOCK_READ, &outData));
 
            cv::Mat cvOut(outData.planes[0].height, outData.planes[0].width, CV_8UC3, outData.planes[0].data,
                          outData.planes[0].pitchBytes);
            imwrite("scaled_" + strBackend + ".png", cvOut);
 
            // Done handling output image, don't forget to unlock it.
            CHECK_STATUS(vpiImageUnlock(output));
        }
    }
    catch (std::exception &e)
    {
        std::cerr << e.what() << std::endl;
        retval = 1;
    }
 
    // Clean up
 
    // Make sure stream is synchronized before destroying the objects
    // that might still be in use.
    vpiStreamSync(stream);
 
    vpiImageDestroy(image);
    vpiImageDestroy(imageNV12);
    vpiImageDestroy(output);
    vpiStreamDestroy(stream);
 
    return retval;
}
 
// vim: ts=8:sw=4:sts=4:et:ai