Skip to content

Conversation

@bresilla
Copy link
Contributor

  • Add separate processing for RGB and Depth camera types with specific capture source and image properties.
  • Introduce new data structures to handle RGB and Depth surface contexts.
  • Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
  • Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
  • Add a Depth array to FRenderRequest struct to store depth information.

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.
@bresilla
Copy link
Contributor Author

Some context: This PR updates the camera component to output depth data in 32FC1, a commonly used format for depth images. In this format, depth values are stored as 32-bit floating-point numbers, representing the distance (in meters) from the camera.

Previously, the implementation was incorrect because the depth image was stored in RGB8 format. In this format, the pixel values—particularly the R channel, which typically holds depth data in Unreal—did not accurately represent depth. Instead, they were just a visual representation rather than meaningful depth values.

@yuokamoto
Copy link
Contributor

Thank you very much!
@james-yoo I added depth camera with your request before. Could you test this pr as well if you have time?
#225 (comment)

Thanks in advance.

@james-yoo
Copy link
Contributor

@yuokamoto
I will check it.

@bresilla
Copy link
Contributor Author

Any update on this??

I am working to add camera segmentation based on Unreal Stencil thingy. Would be cool if this gets reviewed before.

@james-yoo
Copy link
Contributor

@yuokamoto
OK, it seems that this PR output depth data in 32FC1 format. Great!

depth_test.mp4

@yuokamoto
Copy link
Contributor

@yuokamoto OK, it seems that this PR output depth data in 32FC1 format. Great!

depth_test.mp4

Cool thankyou!

@yuokamoto yuokamoto merged commit 3271e58 into rapyuta-robotics:devel Feb 22, 2025
james-yoo pushed a commit to james-yoo/RapyutaSimulationPlugins that referenced this pull request Feb 23, 2025
…s#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.
yuokamoto pushed a commit that referenced this pull request Feb 23, 2025
- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

Co-authored-by: Trim Bresilla <[email protected]>
yuokamoto pushed a commit that referenced this pull request Feb 24, 2025
- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.
yuokamoto added a commit that referenced this pull request Feb 24, 2025
- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

Co-authored-by: Trim Bresilla <[email protected]>
yuokamoto pushed a commit that referenced this pull request Feb 24, 2025
- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.
yuokamoto added a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: Trim Bresilla <[email protected]>
github-actions bot pushed a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: Trim Bresilla <[email protected]>
github-actions bot pushed a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: Trim Bresilla <[email protected]>
github-actions bot pushed a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: Trim Bresilla <[email protected]>
github-actions bot pushed a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: Trim Bresilla <[email protected]>
yuokamoto added a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: yuokamoto <[email protected]>
Co-authored-by: Trim Bresilla <[email protected]>
github-actions bot added a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: yuokamoto <[email protected]>
Co-authored-by: Trim Bresilla <[email protected]>
yuokamoto added a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: yuokamoto <[email protected]>
Co-authored-by: Trim Bresilla <[email protected]>
yuokamoto added a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: yuokamoto <[email protected]>
Co-authored-by: Trim Bresilla <[email protected]>
yuokamoto added a commit that referenced this pull request Feb 24, 2025
* feat: support actual depth image with 32FC1 datatype (#349)

- Add separate processing for RGB and Depth camera types with specific capture source and image properties.
- Introduce new data structures to handle RGB and Depth surface contexts.
- Adjust pixel format and data handling based on camera type, including custom encoding and image data initialization.
- Improve conditions for processing captured data by ensuring proper checks are in place for RGB and Depth data retrieval.
- Add a `Depth` array to `FRenderRequest` struct to store depth information.
- Remove obsolete comments and redundant code sections to improve readability and maintainability.

* update action

---------

Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Co-authored-by: yuokamoto <[email protected]>
Co-authored-by: Trim Bresilla <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

3 participants