Skip to content

New 3D transform algorithm for novel view synthesis. #65

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 6 commits into from
Feb 8, 2023

Conversation

kostarion
Copy link
Contributor

  • Algorithm that applies 3D camera pose transformation to a frame using meshes/pointclouds.
    Each pixel + corresponding depth is being considered as a point in space. The pointcloud or mesh, composed from composition of neighboring points, is rendered to obtain the final image. Returned inpaint mask is rendered in the same fashion. Type of rendering for image and mask can differ.
  • do_prefill parameter for resample methods.

Copy link
Member

@pharmapsychotic pharmapsychotic left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

  • The .pyc files should not be checked in
  • do_prefill can be removed and done with a new BORDER_INFILL since we're supporting for all the transform modes. whatcha think?

Made a pass at getting the new messages to match in better with existing messages. How does this look to you?

message RenderMeshParameters {
  float max_mesh_edge = 1;  // Mesh edges longer that this values are discarded
}

message RenderPointsParameters {
  float radius = 1;             // The radius (in NDC units) of each disk to be rasterized 
  uint32 points_per_pixel = 2;  // Number of points to keep track of per pixel
}

message RenderParameters {
  oneof params {
    RenderMeshParameters mesh = 1;
    RenderPointsParameters points = 2;
  }
}

enum CameraType {
  CAMERA_PERSPECTIVE = 0;   // Perspective camera
  CAMERA_ORTHOGRAPHIC = 1;  // Orthographic camera. Well suited for isometric animations
}

message CameraParameters { 
  CameraType type = 1;
  optional float near_plane = 2; // Nearest plane depth of a rendered frustum
  optional float far_plane = 3;  // Farthest plane depth of a rendered frustum
  optional float fov = 4;        // Camera field of view (in degrees). Must be set for CAMERA_PERSPECTIVE type.
}

message TransformCameraPose {
  CameraParameters camera = 1;     // camera projection settings
  TransformMatrix transform = 2;   // 4x4 transform matrix for the next frame
  RenderParameters image_render = 3;
  RenderParameters mask_render = 4;
}

@sonarqubecloud
Copy link

sonarqubecloud bot commented Feb 8, 2023

Kudos, SonarCloud Quality Gate passed!    Quality Gate passed

Bug A 0 Bugs
Vulnerability A 0 Vulnerabilities
Security Hotspot A 0 Security Hotspots
Code Smell A 0 Code Smells

No Coverage information No Coverage information
No Duplication information No Duplication information

@kostarion kostarion merged commit bfd95e8 into anima_stable Feb 8, 2023
@kostarion kostarion deleted the cut_pow/animation_camera_pose_3d branch February 8, 2023 16:53
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.

2 participants