1 VinModelV3

vin.model_v3.VinModelV3(config)

VIN-Core head for one-step RRI prediction.

VIN v3 focuses on pose encoding, compact voxel evidence, and semidense projection stats, while enforcing fail-fast contracts to avoid silent collapse. It ranks candidates for immediate RRI; bounded rollout values such as \(Q_H\) are separate thesis models trained on rollout replay.

1.1 Attributes

Name Description
pose_encoder_lff Return the LFF sub-encoder when present (else None).

1.2 Methods

Name Description
forward Score candidate poses for one snippet (no diagnostics).
forward_with_debug Run VIN v3 forward pass and return intermediate tensors.
init_bin_values Initialize learnable bin representatives for CORAL expectation.
summarize_vin Summarize VIN v3 inputs/outputs for a single oracle-labeled batch.

1.2.1 forward

vin.model_v3.VinModelV3.forward(
    efm,
    candidate_poses_world_cam,
    reference_pose_world_rig,
    p3d_cameras,
    backbone_out=None,
)

Score candidate poses for one snippet (no diagnostics).

1.2.1.1 Parameters

Name Type Description Default
efm EfmSnippetView | VinSnippetView EFM snippet view or VIN snippet view for the current snippet. required
candidate_poses_world_cam PoseTW Candidate camera poses in world frame. required
reference_pose_world_rig PoseTW Reference rig pose in world frame. required
p3d_cameras PerspectiveCameras Pytorch3D cameras aligned with candidates. required
backbone_out EvlBackboneOutput | None Optional precomputed backbone output. None

1.2.1.2 Returns

Name Type Description
VinPrediction VinPrediction containing ordinal logits, expected scores, and
VinPrediction validity masks for each candidate.

1.2.2 forward_with_debug

vin.model_v3.VinModelV3.forward_with_debug(
    efm,
    candidate_poses_world_cam,
    reference_pose_world_rig,
    p3d_cameras,
    backbone_out=None,
)

Run VIN v3 forward pass and return intermediate tensors.

1.2.2.1 Parameters

Name Type Description Default
efm EfmSnippetView | VinSnippetView EFM snippet view or VIN snippet view for the current snippet. required
candidate_poses_world_cam PoseTW Candidate camera poses in world frame. required
reference_pose_world_rig PoseTW Reference rig pose in world frame. required
p3d_cameras PerspectiveCameras Pytorch3D cameras aligned with candidates. required
backbone_out EvlBackboneOutput | None Optional precomputed backbone output. None

1.2.2.2 Returns

Name Type Description
tuple[VinPrediction, VinV3ForwardDiagnostics] Tuple of (VinPrediction, VinV3ForwardDiagnostics).

1.2.3 init_bin_values

vin.model_v3.VinModelV3.init_bin_values(values, *, overwrite=False)

Initialize learnable bin representatives for CORAL expectation.

1.2.3.1 Parameters

Name Type Description Default
values Tensor Tensor["K"] target bin representatives (e.g., bin means). required
overwrite bool If True, overwrite existing bin values. False

1.2.4 summarize_vin

vin.model_v3.VinModelV3.summarize_vin(
    batch,
    *,
    include_torchsummary=True,
    torchsummary_depth=3,
)

Summarize VIN v3 inputs/outputs for a single oracle-labeled batch.

This is intended for quick sanity checks when validating the streamlined baseline against sweep-derived expectations.

1.2.4.1 Parameters

Name Type Description Default
batch VinOracleBatch Oracle-labeled batch to inspect. required
include_torchsummary bool Whether to include a torchsummary block. True
torchsummary_depth int Depth for torchsummary. 3

1.2.4.2 Returns

Name Type Description
str Human-readable summary string.