imageprocessing/geometry/pose/docs/usecase1.md

Go to the documentation of this file.

# Copyright (c) 2019-2020, NVIDIA CORPORATION.  All rights reserved.

@page imageprocessing_geometry_pose_usecase1 Pose estimation

@note SW Release Applicability: This tutorial is applicable to modules in both **NVIDIA DriveWorks** and **NVIDIA DRIVE Software** releases.

The tutorial shows the general structure of a program that uses the PnP pose estimator to determine the camera pose.

#### Initialize the pose estimator

We select a conservative number of ransac and non-linear optimizer iterations (10 and 10).
`contextHandle` is assumed to be previously initialized `::dwContextHandle_t`.

```{.cpp}
    dwPnP_initialize(&pnp, 10, 10, contextHandle);

```
#### Obtain matches

```{.cpp}
    dwVector3f worldPoints[BUFFER_SIZE];
    dwVector2f imagePoints[BUFFER_SIZE];
    size_t pointCount;

    // The application must obtain the matches and fill the above three variables

```

#### Undistort feature points

`cameraModel` is assumed to be previously initialized `::dwCameraModelHandle_t`

```{.cpp}
    dwVector2f rays[BUFFER_SIZE];

    for(size_t i = 0; i < pointCount; ++i)
    {
        const auto& point = imagePoints[i];
        auto& ray = rays[i];
        dwCameraModel_pixel2Ray(&ray.x, &ray.y, &ray.z, point.x, point.y, cameraModel);
    }

```

#### Estimate pose

```{.cpp}
    dwTransformation3f worldToCamera;
    dwPnP_solve(&worldToCamera, pointCount, rays, worldPoints, pnp);

```
#### Finally, free previously allocated memory.

```{.cpp}
    dwPnP_release(pnp);
```