Stereo Disparity

Overview

The Stereo Disparity application receives left and right stereo pair images and returns the disparity between them, which is a function of image depth. The result is saved as an image file to disk. If available, it'll also output the corresponding confidence map.

Instructions

The command line parameters are:

<backend> <left image> <right image>

where

• backend: either cuda, ofa or ofa-pva-vic; it defines the backend that will perform the processing. ofa-pva-vic and cuda allow output of the confidence map in addition to the disparity.
• left image: left input image of a rectified stereo pair, it accepts png, jpeg and possibly others.
• right image: right input image of a stereo pair.

Here's one example:

• C++

  ./vpi_sample_02_stereo_disparity cuda ../assets/chair_stereo_left.png ../assets/chair_stereo_right.png

• Python

  python3 main.py cuda ../assets/chair_stereo_left.png ../assets/chair_stereo_right.png

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

The Python version of this sample also allow for setting various additional parameters, as well as additional input image extensions and to turn on verbose mode. The following command-line arguments can be passed to the Python sample:

• Python

  python3 main.py <backend> <left image> <right image> --width W --height H --downscale D --window_size WIN
      --skip_confidence --conf_threshold T --conf_type absolute/relative -p1 P1 -p2 P2 --p2_alpha P2alpha
      --uniqueness U --skip_diagonal --num_passes N --min_disparity MIN --max_disparity MAX --output_mode 0/1/2
      -v/--verbose

  where the additional optional arguments are:
• width: set the width W when passing ".raw" input images
• height: set the height H when passing ".raw" input images
• downscale: to set the output downscale factor as D
• window_size: to set the median filter window size as WIN
• skip_confidence: to avoid calculating confidence and applying it as a mask
• conf_threshold: set the confidence threshold as T
• conf_type: set the confidence type as either absolute or relative
• p1: set p1 penalty as P1
• p2: set p2 penalty as P2
• p2_alpha: set p2Alpha adaptive penalty as P2alpha
• uniqueness: set uniqueness as U
• skip_diagonal: to avoid using diagonal paths in CUDA or OFA backends
• num_passes: to set the number of passes N in OFA backends
• min_disparity: to set the minimum disparity MIN in CUDA backend
• max_disparity: to set the maximum disparity MAX in backends
• output_mode: 0 for colored output, 1 for grayscale and 2 for raw binary
• verbose: to turn on verbose mode To understand in detail each of these aditional arguments related to the stereo disparity algorithm, please read the corresponding documentation.

Results

Left input image	Right input image
Stereo disparity	Confidence map

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
import cv2
 
 
def read_raw_file(fpath, resize_to=None, verbose=False):
    try:
        if verbose:
            print(f'I Reading: {fpath}', end=' ', flush=True)
        f = open(fpath, 'rb')
        np_arr = np.fromfile(f, dtype=np.uint16, count=-1)
        f.close()
        if verbose:
            print(f'done!\nI Raw array: shape: {np_arr.shape} dtype: {np_arr.dtype}')
        if resize_to is not None:
            np_arr = np_arr.reshape(resize_to, order='C')
        if verbose:
            print(f'I Reshaped array: shape: {np_arr.shape} dtype: {np_arr.dtype}')
        pil_img = Image.fromarray(np_arr, mode="I;16L")
        return pil_img
    except:
        raise ValueError(f'E Cannot process raw input: {fpath}')
 
 
def process_arguments():
    parser = ArgumentParser()
 
    parser.add_argument('backend', choices=['cuda','ofa','ofa-pva-vic'],
                        help='Backend to be used for processing')
    parser.add_argument('left', help='Rectified left input image from a stereo pair')
    parser.add_argument('right', help='Rectified right input image from a stereo pair')
    parser.add_argument('--width', default=-1, type=int, help='Input width for raw input files')
    parser.add_argument('--height', default=-1, type=int, help='Input height for raw input files')
    parser.add_argument('--downscale', default=1, type=int, help='Output downscale factor')
    parser.add_argument('--window_size', default=5, type=int, help='Median filter window size')
    parser.add_argument('--skip_confidence', default=False, action='store_true', help='Do not calculate confidence')
    parser.add_argument('--conf_threshold', default=32767, type=int, help='Confidence threshold')
    parser.add_argument('--conf_type', default='best', choices=['best', 'absolute', 'relative', 'inference'],
                        help='Computation type to produce the confidence output. Default will pick best option given backend.')
    parser.add_argument('-p1', default=3, type=int, help='Penalty P1 on small disparities')
    parser.add_argument('-p2', default=48, type=int, help='Penalty P2 on large disparities')
    parser.add_argument('--p2_alpha', default=0, type=int, help='Alpha for adaptive P2 Penalty')
    parser.add_argument('--uniqueness', default=-1, type=float, help='Uniqueness ratio')
    parser.add_argument('--skip_diagonal', default=False, action='store_true', help='Do not use diagonal paths')
    parser.add_argument('--num_passes', default=3, type=int, help='Number of passes')
    parser.add_argument('--min_disparity', default=0, type=int, help='Minimum disparity')
    parser.add_argument('--max_disparity', default=256, type=int, help='Maximum disparity')
    parser.add_argument('--output_mode', default=0, type=int, help='0: color; 1: grayscale; 2: raw binary')
    parser.add_argument('-v', '--verbose', default=False, action='store_true', help='Verbose mode')
 
    return parser.parse_args()
 
 
def main():
    args = process_arguments()
 
    scale = 1 # pixel value scaling factor when loading input
 
    if args.backend == 'cuda':
        backend = vpi.Backend.CUDA
    elif args.backend == 'ofa':
        backend = vpi.Backend.OFA
    elif args.backend == 'ofa-pva-vic':
        backend = vpi.Backend.OFA|vpi.Backend.PVA|vpi.Backend.VIC
    else:
        raise ValueError(f'E Invalid backend: {args.backend}')
 
    conftype = None
    if args.conf_type == 'best':
        conftype = vpi.ConfidenceType.INFERENCE if args.backend == 'ofa-pva-vic' else vpi.ConfidenceType.ABSOLUTE
    elif args.conf_type == 'absolute':
        conftype = vpi.ConfidenceType.ABSOLUTE
    elif args.conf_type == 'relative':
        conftype = vpi.ConfidenceType.RELATIVE
    elif args.conf_type == 'inference':
        conftype = vpi.ConfidenceType.INFERENCE
    else:
        raise ValueError(f'E Invalid confidence type: {args.conf_type}')
 
    minDisparity = args.min_disparity
    maxDisparity = args.max_disparity
    includeDiagonals = not args.skip_diagonal
    numPasses = args.num_passes
    calcConf = not args.skip_confidence
    downscale = args.downscale
    windowSize = args.window_size
    quality = 6
 
    if args.verbose:
        print(f'I Backend: {backend}\nI Left image: {args.left}\nI Right image: {args.right}\n'
              f'I Disparities (min, max): {(minDisparity, maxDisparity)}\n'
              f'I Input scale factor: {scale}\nI Output downscale factor: {downscale}\n'
              f'I Window size: {windowSize}\nI Quality: {quality}\n'
              f'I Calculate confidence: {calcConf}\nI Confidence threshold: {args.conf_threshold}\n'
              f'I Confidence type: {conftype}\nI Uniqueness ratio: {args.uniqueness}\n'
              f'I Penalty P1: {args.p1}\nI Penalty P2: {args.p2}\nI Adaptive P2 alpha: {args.p2_alpha}\n'
              f'I Include diagonals: {includeDiagonals}\nI Number of passes: {numPasses}\n'
              f'I Output mode: {args.output_mode}\nI Verbose: {args.verbose}\n'
              , end='', flush=True)
 
    if 'raw' in args.left:
        pil_left = read_raw_file(args.left, resize_to=[args.height, args.width], verbose=args.verbose)
        np_left = np.asarray(pil_left)
    else:
        try:
            pil_left = Image.open(args.left)
            if pil_left.mode == 'I':
                np_left = np.asarray(pil_left).astype(np.int16)
            else:
                np_left = np.asarray(pil_left)
        except:
            raise ValueError(f'E Cannot open left input image: {args.left}')
 
    if 'raw' in args.right:
        pil_right = read_raw_file(args.right, resize_to=[args.height, args.width], verbose=args.verbose)
        np_right = np.asarray(pil_right)
    else:
        try:
            pil_right = Image.open(args.right)
            if pil_right.mode == 'I':
                np_right = np.asarray(pil_right).astype(np.int16)
            else:
                np_right = np.asarray(pil_right)
        except:
            raise ValueError(f'E Cannot open right input image: {args.right}')
 
    # Streams for left and right independent pre-processing
    streamLeft = vpi.Stream()
    streamRight = vpi.Stream()
 
    # Load input into a vpi.Image and convert it to grayscale, 16bpp
    with vpi.Backend.CUDA:
        with streamLeft:
            left = vpi.asimage(np_left).convert(vpi.Format.Y16_ER, scale=scale)
        with streamRight:
            right = vpi.asimage(np_right).convert(vpi.Format.Y16_ER, scale=scale)
 
    # Preprocess input
    # Block linear format is needed for ofa backends
    # We use VIC backend for the format conversion because it is low power
    if args.backend in {'ofa-pva-vic', 'ofa'}:
        if args.verbose:
            print(f'W {args.backend} forces to convert input images to block linear', flush=True)
        with vpi.Backend.VIC:
            with streamLeft:
                left = left.convert(vpi.Format.Y16_ER_BL)
            with streamRight:
                right = right.convert(vpi.Format.Y16_ER_BL)
 
    if args.verbose:
        print(f'I Input left image: {left.size} {left.format}\n'
              f'I Input right image: {right.size} {right.format}', flush=True)
 
    confidenceU16 = None
 
    if calcConf:
        if args.backend not in {'cuda', 'ofa-pva-vic'}:
            # Only CUDA and OFA-PVA-VIC support confidence map
            calcConf = False
            if args.verbose:
                print(f'W {args.backend} does not allow to calculate confidence', flush=True)
 
 
    outWidth = (left.size[0] + downscale - 1) // downscale
    outHeight = (left.size[1] + downscale - 1) // downscale
 
    if calcConf:
        confidenceU16 = vpi.Image((outWidth, outHeight), vpi.Format.U16)
 
    # Use stream left to consolidate actual stereo processing
    streamStereo = streamLeft
 
    if args.backend == 'ofa-pva-vic' and maxDisparity not in {128, 256}:
        maxDisparity = 128 if (maxDisparity // 128) < 1 else 256
        if args.verbose:
            print(f'W {args.backend} only supports 128 or 256 maxDisparity. Overriding to {maxDisparity}', flush=True)
 
    if args.verbose:
        if 'ofa' not in args.backend:
            print('W Ignoring P2 alpha and number of passes since not an OFA backend', flush=True)
        if args.backend != 'cuda':
            print('W Ignoring uniqueness since not a CUDA backend', flush=True)
        print('I Estimating stereo disparity ... ', end='', flush=True)
 
    # Estimate stereo disparity.
    with streamStereo, backend:
        disparityS16 = vpi.stereodisp(left, right, downscale=downscale, out_confmap=confidenceU16,
                                      window=windowSize, maxdisp=maxDisparity, confthreshold=args.conf_threshold,
                                      quality=quality, conftype=conftype, mindisp=minDisparity,
                                      p1=args.p1, p2=args.p2, p2alpha=args.p2_alpha, uniqueness=args.uniqueness,
                                      includediagonals=includeDiagonals, numpasses=numPasses)
 
    if args.verbose:
        print('done!\nI Post-processing ... ', end='', flush=True)
 
    # Postprocess results and save them to disk
    with streamStereo, vpi.Backend.CUDA:
        # Some backends outputs disparities in block-linear format, we must convert them to
        # pitch-linear for consistency with other backends.
        if disparityS16.format == vpi.Format.S16_BL:
            disparityS16 = disparityS16.convert(vpi.Format.S16, backend=vpi.Backend.VIC)
 
        # Scale disparity and confidence map so that values like between 0 and 255.
 
        # Disparities are in Q10.5 format, so to map it to float, it gets
        # divided by 32. Then the resulting disparity range, from 0 to
        # stereo.maxDisparity gets mapped to 0-255 for proper output.
        # Copy disparity values back to the CPU.
        disparityU8 = disparityS16.convert(vpi.Format.U8, scale=255.0/(32*maxDisparity)).cpu()
 
        # Apply JET colormap to turn the disparities into color, reddish hues
        # represent objects closer to the camera, blueish are farther away.
        disparityColor = cv2.applyColorMap(disparityU8, cv2.COLORMAP_JET)
 
        # Converts to RGB for output with PIL.
        disparityColor = cv2.cvtColor(disparityColor, cv2.COLOR_BGR2RGB)
 
        if calcConf:
            confidenceU8 = confidenceU16.convert(vpi.Format.U8, scale=255.0/65535).cpu()
 
            # When pixel confidence is 0, its color in the disparity is black.
            mask = cv2.threshold(confidenceU8, 1, 255, cv2.THRESH_BINARY)[1]
            mask = cv2.cvtColor(mask, cv2.COLOR_GRAY2BGR)
            disparityColor = cv2.bitwise_and(disparityColor, mask)
 
    fext = '.raw' if args.output_mode == 2 else '.png'
 
    disparity_fname = f'disparity_python{sys.version_info[0]}_{args.backend}' + fext
    confidence_fname = f'confidence_python{sys.version_info[0]}_{args.backend}' + fext
 
    if args.verbose:
        print(f'done!\nI Disparity output: {disparity_fname}', flush=True)
        if calcConf:
            print(f'I Confidence output: {confidence_fname}', flush=True)
 
    # Save results to disk.
    try:
        if args.output_mode == 0:
            Image.fromarray(disparityColor).save(disparity_fname)
            if args.verbose:
                print(f'I Output disparity image: {disparityColor.shape} '
                      f'{disparityColor.dtype}', flush=True)
        elif args.output_mode == 1:
            Image.fromarray(disparityU8).save(disparity_fname)
            if args.verbose:
                print(f'I Output disparity image: {disparityU8.shape} '
                      f'{disparityU8.dtype}', flush=True)
        elif args.output_mode == 2:
            disparityS16.cpu().tofile(disparity_fname)
            if args.verbose:
                print(f'I Output disparity image: {disparityS16.size} '
                      f'{disparityS16.format}', flush=True)
 
        if calcConf:
            if args.output_mode == 0 or args.output_mode == 1:
                Image.fromarray(confidenceU8).save(confidence_fname)
                if args.verbose:
                    print(f'I Output confidence image: {confidenceU8.shape} '
                          f'{confidenceU8.dtype}', flush=True)
            else:
                confidenceU16.cpu().tofile(confidence_fname)
                if args.verbose:
                    print(f'I Output confidence image: {confidenceU16.size} '
                          f'{confidenceU16.format}', flush=True)
 
    except:
        raise ValueError(f'E Cannot write outputs: {disparity_fname}, {confidence_fname}\n'
                         f'E Using output mode: {args.output_mode}')
 
 
if __name__ == '__main__':
    main()

#include <opencv2/core/version.hpp>
#if CV_MAJOR_VERSION >= 3
#    include <opencv2/imgcodecs.hpp>
#else
#    include <opencv2/contrib/contrib.hpp> // for colormap
#    include <opencv2/highgui/highgui.hpp>
#endif
 
#include <opencv2/imgproc/imgproc.hpp>
#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 <vpi/algo/StereoDisparity.h>
 
#include <cstring>
#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 << "line " << __LINE__ << " " << 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 cvImageLeft, cvImageRight;
 
    // VPI objects that will be used
    VPIImage inLeft        = NULL;
    VPIImage inRight       = NULL;
    VPIImage tmpLeft       = NULL;
    VPIImage tmpRight      = NULL;
    VPIImage stereoLeft    = NULL;
    VPIImage stereoRight   = NULL;
    VPIImage disparity     = NULL;
    VPIImage confidenceMap = NULL;
    VPIStream stream       = NULL;
    VPIPayload stereo      = NULL;
 
    int retval = 0;
 
    try
    {
        // =============================
        // Parse command line parameters
 
        if (argc != 4)
        {
            throw std::runtime_error(std::string("Usage: ") + argv[0] +
                                     " <cuda|ofa|ofa-pva-vic> <left image> <right image>");
        }
 
        std::string strBackend       = argv[1];
        std::string strLeftFileName  = argv[2];
        std::string strRightFileName = argv[3];
 
        uint64_t backends;
 
        if (strBackend == "cuda")
        {
            backends = VPI_BACKEND_CUDA;
        }
        else if (strBackend == "ofa")
        {
            backends = VPI_BACKEND_OFA;
        }
        else if (strBackend == "ofa-pva-vic")
        {
            backends = VPI_BACKEND_OFA | VPI_BACKEND_PVA | VPI_BACKEND_VIC;
        }
        else
        {
            throw std::runtime_error("Backend '" + strBackend +
                                     "' not recognized, it must be either cuda, ofa or ofa-pva-vic.");
        }
 
        // =====================
        // Load the input images
        cvImageLeft = cv::imread(strLeftFileName);
        if (cvImageLeft.empty())
        {
            throw std::runtime_error("Can't open '" + strLeftFileName + "'");
        }
 
        cvImageRight = cv::imread(strRightFileName);
        if (cvImageRight.empty())
        {
            throw std::runtime_error("Can't open '" + strRightFileName + "'");
        }
 
        // =================================
        // Allocate all VPI resources needed
 
        int32_t inputWidth  = cvImageLeft.cols;
        int32_t inputHeight = cvImageLeft.rows;
 
        // Create the stream that will be used for processing.
        CHECK_STATUS(vpiStreamCreate(0, &stream));
 
        // We now wrap the loaded images 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(vpiImageCreateWrapperOpenCVMat(cvImageLeft, 0, &inLeft));
        CHECK_STATUS(vpiImageCreateWrapperOpenCVMat(cvImageRight, 0, &inRight));
 
        // Format conversion parameters needed for input pre-processing
        VPIConvertImageFormatParams convParams;
        CHECK_STATUS(vpiInitConvertImageFormatParams(&convParams));
 
        // Initialize default parameters
        VPIStereoDisparityEstimatorCreationParams createParams;
        CHECK_STATUS(vpiInitStereoDisparityEstimatorCreationParams(&createParams));
 
        // Select max disparity that works well for the chair_stereo_{left,right}_1920.png files
        createParams.maxDisparity = 256;
 
        // Default format and size for input stereo pair (some backends require adjustments, see below)
        VPIImageFormat stereoFormat = VPI_IMAGE_FORMAT_Y8_ER;
 
        int stereoWidth  = inputWidth;
        int stereoHeight = inputHeight;
 
        // Default format and size for output
        VPIImageFormat disparityFormat = VPI_IMAGE_FORMAT_S16;
 
        int outputWidth  = inputWidth;
        int outputHeight = inputHeight;
 
        // Override some backend-dependent parameters
        if (strBackend.find("ofa") != std::string::npos)
        {
            // Implementations using OFA require BL input
            stereoFormat = VPI_IMAGE_FORMAT_Y8_ER_BL;
 
            if (strBackend == "ofa")
            {
                // when using OFA alone, output must also be BL
                disparityFormat = VPI_IMAGE_FORMAT_S16_BL;
            }
 
            // Using downscale factor with OFA improves performance
            createParams.downscaleFactor = 2;
            outputWidth  = (inputWidth + createParams.downscaleFactor - 1) / createParams.downscaleFactor;
            outputHeight = (inputHeight + createParams.downscaleFactor - 1) / createParams.downscaleFactor;
 
            // Output width including downscaleFactor must be at least max(64, maxDisparity/downscaleFactor) when the
            // OFA+PVA+VIC backend is used
            if (strBackend.find("pva") != std::string::npos)
            {
                int minWidth = std::max(createParams.maxDisparity / createParams.downscaleFactor, outputWidth);
                outputWidth  = std::max(64, minWidth);
                outputHeight = (inputHeight * outputWidth) / inputWidth;
                stereoWidth  = outputWidth * createParams.downscaleFactor;
                stereoHeight = outputHeight * createParams.downscaleFactor;
            }
        }
 
        // Create the payload for Stereo Disparity algorithm.
        // Payload is created before the image objects so that non-supported backends can be trapped with an error.
        CHECK_STATUS(vpiCreateStereoDisparityEstimator(backends, stereoWidth, stereoHeight, stereoFormat, &createParams,
                                                       &stereo));
 
        // Create the output image where the disparity map will be stored.
        CHECK_STATUS(vpiImageCreate(outputWidth, outputHeight, disparityFormat, 0, &disparity));
 
        // Create the input stereo images
        CHECK_STATUS(vpiImageCreate(stereoWidth, stereoHeight, stereoFormat, 0, &stereoLeft));
        CHECK_STATUS(vpiImageCreate(stereoWidth, stereoHeight, stereoFormat, 0, &stereoRight));
 
        // Create the confidence image if the backend can support it
        if (strBackend == "ofa-pva-vic" || strBackend == "cuda")
        {
            CHECK_STATUS(vpiImageCreate(outputWidth, outputHeight, VPI_IMAGE_FORMAT_U16, 0, &confidenceMap));
        }
 
        // If a rescale of the input is required, create temporary images for the initial format conversion.
        bool const isRescaleRequired = (stereoWidth != inputWidth) || (stereoHeight != inputHeight);
        if (isRescaleRequired)
        {
            CHECK_STATUS(vpiImageCreate(inputWidth, inputHeight, stereoFormat, 0, &tmpLeft));
            CHECK_STATUS(vpiImageCreate(inputWidth, inputHeight, stereoFormat, 0, &tmpRight));
        }
 
        // ================
        // Processing stage
 
        // Start with default parameters, and override some values depending on what backend is used.
        VPIStereoDisparityEstimatorParams submitParams;
        CHECK_STATUS(vpiInitStereoDisparityEstimatorParams(&submitParams));
        if (strBackend == "ofa-pva-vic")
        {
            // The INFERENCE confidence type achieves better performance with OFA+PVA+VIC backend. The only tradeoff is
            // that the deep-learning based confidence map is not easily expressed as a function of left and right
            // disparity estimates, in contrast to ABSOLUTE or RELATIVE confidence type.
            submitParams.confidenceType = VPI_STEREO_CONFIDENCE_INFERENCE;
        }
        else if (strBackend == "cuda")
        {
            // The chair_stereo_{left,right}_1920.png inputs benefit from a higher confidence threshold with CUDA
            submitParams.confidenceThreshold = UINT16_MAX - 10000;
        }
 
        // -----------------
        // Pre-process input
        if (isRescaleRequired)
        {
            // We require a conversion with CUDA only because we loaded the images in the default BGR format from OpenCV
            // and the VIC backend does not support 3-channel RGB/BGR image formats.
            // Alternatively, we could load grayscale images and handle the conversion+rescale in one operation on VIC.
 
            // Convert opencv input to grayscale format using CUDA
            CHECK_STATUS(vpiSubmitConvertImageFormat(stream, VPI_BACKEND_CUDA, inLeft, tmpLeft, &convParams));
            CHECK_STATUS(vpiSubmitConvertImageFormat(stream, VPI_BACKEND_CUDA, inRight, tmpRight, &convParams));
 
            // Rescale on VIC
            CHECK_STATUS(
                vpiSubmitRescale(stream, VPI_BACKEND_VIC, tmpLeft, stereoLeft, VPI_INTERP_LINEAR, VPI_BORDER_CLAMP, 0));
            CHECK_STATUS(vpiSubmitRescale(stream, VPI_BACKEND_VIC, tmpRight, stereoRight, VPI_INTERP_LINEAR,
                                          VPI_BORDER_CLAMP, 0));
        }
        else
        {
            // Convert opencv input to grayscale format using CUDA
            CHECK_STATUS(vpiSubmitConvertImageFormat(stream, VPI_BACKEND_CUDA, inLeft, stereoLeft, &convParams));
            CHECK_STATUS(vpiSubmitConvertImageFormat(stream, VPI_BACKEND_CUDA, inRight, stereoRight, &convParams));
        }
 
        // ------------------------------
        // Do stereo disparity estimation
 
        // Submit it with the input and output images
        CHECK_STATUS(vpiSubmitStereoDisparityEstimator(stream, backends, stereo, stereoLeft, stereoRight, disparity,
                                                       confidenceMap, &submitParams));
 
        // Wait until the algorithm finishes processing
        CHECK_STATUS(vpiStreamSync(stream));
 
        // ========================================
        // Output pre-processing and saving to disk
        // Lock output to retrieve its data on cpu memory
        VPIImageData data;
        CHECK_STATUS(vpiImageLockData(disparity, VPI_LOCK_READ, VPI_IMAGE_BUFFER_HOST_PITCH_LINEAR, &data));
 
        // Make an OpenCV matrix out of this image
        cv::Mat cvDisparity;
        CHECK_STATUS(vpiImageDataExportOpenCVMat(data, &cvDisparity));
 
        // Scale result and write it to disk. Disparities are in Q10.5 format,
        // so to map it to float, it gets divided by 32. Then the resulting disparity range,
        // from 0 to maxDisparity gets mapped to 0-255 for proper output.
        cvDisparity.convertTo(cvDisparity, CV_8UC1, 255.0 / (32 * createParams.maxDisparity), 0);
 
        // Apply JET colormap to turn the disparities into color.
        // Reddish hues represent objects closer to the camera, blueish are farther away.
        cv::Mat cvDisparityColor;
        applyColorMap(cvDisparity, cvDisparityColor, cv::COLORMAP_JET);
 
        // Done handling output, don't forget to unlock it.
        CHECK_STATUS(vpiImageUnlock(disparity));
 
        // If we have a confidence map, adjust it for display and write it to disk too.
        if (confidenceMap)
        {
            // Lock the image data and export to cv::Mat
            VPIImageData data;
            CHECK_STATUS(vpiImageLockData(confidenceMap, VPI_LOCK_READ, VPI_IMAGE_BUFFER_HOST_PITCH_LINEAR, &data));
            cv::Mat cvConfidence;
            CHECK_STATUS(vpiImageDataExportOpenCVMat(data, &cvConfidence));
 
            // Confidence map varies from 0 to 65535, we scale it to [0-255].
            cvConfidence.convertTo(cvConfidence, CV_8UC1, 255.0 / 65535, 0);
            imwrite("confidence_" + strBackend + ".png", cvConfidence);
 
            CHECK_STATUS(vpiImageUnlock(confidenceMap));
 
            // When pixel confidence is 0, we would like its color in the disparity image to be black.
            cv::Mat cvMask;
            threshold(cvConfidence, cvMask, 1, 255, cv::THRESH_BINARY);
            cvtColor(cvMask, cvMask, cv::COLOR_GRAY2BGR);
            bitwise_and(cvDisparityColor, cvMask, cvDisparityColor);
        }
 
        imwrite("disparity_" + strBackend + ".png", cvDisparityColor);
    }
    catch (std::exception &e)
    {
        std::cerr << e.what() << std::endl;
        retval = 1;
    }
 
    // ========
    // Clean up
 
    // Destroying stream first makes sure that all work submitted to
    // it is finished.
    vpiStreamDestroy(stream);
 
    // Only then we can destroy the other objects, as we're sure they
    // aren't being used anymore.
 
    vpiImageDestroy(inLeft);
    vpiImageDestroy(inRight);
    vpiImageDestroy(tmpLeft);
    vpiImageDestroy(tmpRight);
    vpiImageDestroy(stereoLeft);
    vpiImageDestroy(stereoRight);
    vpiImageDestroy(confidenceMap);
    vpiImageDestroy(disparity);
    vpiPayloadDestroy(stereo);
 
    return retval;
}