- Unified the canonical local frame for generated geometries (Plane, Depth Plane, Relief Plane, Cutout, Depth Cutout) to align with Blender's object coordinate system, setting X as horizontal, Z as up, and +Y as depth.
- Removed the `Depth Axis` parameter from `O Image Cutout` and `O Image Depth Cutout`, simplifying the geometry generation process.
- Updated the depth direction defaults to use +Y instead of -Z for `O Image Relief Plane` and Depth Symmetry.
- Adjusted the geometry nodes to ensure thickness, projection, and shell displacements operate along the Y-axis.
- Implemented new capabilities for handling object space normals to align with the new canonical frame.
- Enhanced the Depth Symmetry functionality to mirror across the canonical Y-axis and perform a final rotation to maintain the correct orientation.
- Added tests to validate the new coordinate system and ensure compatibility with existing assets.
- Added validation documentation for AI model tiers.
- Introduced `background.py` for background removal logic, including inference functions for alpha and foreground extraction.
- Created `onnx_birefnet.py` to handle BiRefNet model sessions and alpha inference.
- Developed tests for background models, ensuring proper dispatch and session reuse.
- Added tests for BiRefNet ONNX model to validate alpha output and error handling for invalid values.
- Implemented model export scripts for BiRefNet and HAT Sharper, ensuring proper weight compression and licensing.
- Created `deform_sampling.py` for deformable convolution operations used in BiRefNet.
- Added utility functions for compressing ONNX weights without altering graph arithmetic.
- Introduced a reference plane node for depth field adjustments in geometry processing.
