1 vin.encoders.trajectory

vin.encoders.trajectory

Trajectory encoders for EFM snippet rig poses.

This module owns a trajectory encoder that reuses the VIN candidate pose encoder per frame and then pools the encoded sequence into optional global context features.

1.1 Classes

Name Description
TrajectoryEncoder Encode EFM trajectory poses with an R6D + LFF pose encoder.
TrajectoryEncoderConfig Configure per-frame pose encoding and sequence reduction.
TrajectoryEncodingOutput Trajectory encoding outputs.

1.1.1 TrajectoryEncoder

vin.encoders.trajectory.TrajectoryEncoder(config)

Encode EFM trajectory poses with an R6D + LFF pose encoder.

1.1.1.1 Attributes

Name Description
out_dim Return the per-frame and pooled trajectory embedding width.

1.1.1.2 Methods

Name Description
encode_poses Encode a batch of trajectory poses.
forward Encode a trajectory’s world←rig poses.
1.1.1.2.1 encode_poses
vin.encoders.trajectory.TrajectoryEncoder.encode_poses(poses)

Encode a batch of trajectory poses.

1.1.1.2.1.1 Parameters
Name Type Description Default
poses PoseTW PoseTW["B F 12"] or PoseTW["F 12"] trajectory poses. required
1.1.1.2.1.2 Returns
Name Type Description
TrajectoryEncodingOutput TrajectoryEncodingOutput with per-frame encodings and pooled embedding.
1.1.1.2.2 forward
vin.encoders.trajectory.TrajectoryEncoder.forward(trajectory)

Encode a trajectory’s world←rig poses.

1.1.1.2.2.1 Parameters
Name Type Description Default
trajectory EfmTrajectoryView aria_nbv.data_handling.ase_efm.views.EfmTrajectoryView. required
1.1.1.2.2.2 Returns
Name Type Description
TrajectoryEncodingOutput TrajectoryEncodingOutput with per-frame encodings and pooled embedding.

1.1.2 TrajectoryEncoderConfig

vin.encoders.trajectory.TrajectoryEncoderConfig()

Configure per-frame pose encoding and sequence reduction.

1.1.2.1 Attributes

Name Description
target_type Factory target for aria_nbv.configs.base.BaseConfig.setup_target.
pose_encoder Pose encoder configuration (R6D + LFF).
pool_mode Pooling mode over frames: mean, final, or none.

1.1.3 TrajectoryEncodingOutput

vin.encoders.trajectory.TrajectoryEncodingOutput(per_frame, pooled)

Trajectory encoding outputs.

1.1.3.1 Attributes

Name Type Description
per_frame PoseEncodingOutput PoseEncodingOutput for each frame.
pooled Tensor | None Tensor["B E", float32] pooled trajectory embedding (or None).
Back to top