Struct CameraEmbedding 

pub struct CameraEmbedding { /* private fields */ }

MLP that lifts a flat camera-pose vector to the time-embedding dimension.

Implementations

impl CameraEmbedding

pub fn new( vs: VarBuilder<'_>, pose_dim: usize, embed_dim: usize, ) -> Result<Self>

Create a new camera embedding MLP.

• pose_dim: input dimension (typically 12 for a flattened 4×3 matrix).
• embed_dim: output dimension (should match the time-embedding dim).

Trait Implementations

impl Debug for CameraEmbedding

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Module for CameraEmbedding

fn forward(&self, pose: &Tensor) -> Result<Tensor>