anchors

Generate axis-aligned anchors for a single class over a BEV feature map.

A port of OpenPCDet's AnchorGenerator with align_center=False: anchor centers are placed on a grid spanning point_cloud_range (endpoints inclusive), then anchor_bottom_heights are lifted by half the box height to box centers.

Parameters:

• point_cloud_range (Sequence[float]) –

  Range \((x_\min, y_\min, z_\min, x_\max, y_\max, z_\max)\).

• feature_map_size (Tuple[int, int]) –

  BEV feature map size \((n_x, n_y)\).

• anchor_sizes (Sequence[Sequence[float]]) –

  Box sizes \((d_x, d_y, d_z)\), one row per size template.

• anchor_rotations (Sequence[float]) –

  Yaw angles (radians).

• anchor_bottom_heights (Sequence[float]) –

  Anchor bottom \(z\) per height template.

• dtype (dtype, default: float32 ) –

  Anchor dtype.

Returns:

• Tensor –

  Anchors \((1, n_y, n_x, n_\text{size}, n_\text{rot}, 7)\) as \((x, y, z, dx, dy, dz, \theta)\).

Assign classification and box-regression targets to a single class group of axis-aligned anchors.

Each anchor is matched to the ground-truth box of highest IoU: IoU \(\ge\) matched_threshold makes it a positive carrying that box's label, IoU \(<\) unmatched_threshold makes it background, and anything in between is ignored. Each ground-truth box additionally force-matches its single highest-IoU anchor, so a box with no anchor above threshold still receives one positive. Positive anchors' regression targets are the residual encoding of their matched box against the anchor (the inverse of decode_box_residuals); the columns a box carries beyond its geometry (a velocity) are copied into the targets as they are.

Callers with several class groups (one anchor set per class) invoke this once per group with that class's anchors, ground truth, and thresholds, then concatenate the results. With gt_batch (the scene index of every box) and batch_size, one call matches the anchors in every scene of a batch: the IoU runs once over the whole batch and the outputs gain a leading scene dimension.

Parameters:

• anchors (Tensor) –

  Anchors \((x, y, z, d_x, d_y, d_z, \theta)\) for one class group, shape \((A, 7)\).

• gt_boxes (Tensor) –

  Ground-truth boxes \((c_x, c_y, c_z, d_x, d_y, d_z, \theta, \ldots)\), shape \((G, 7 + C)\).

• gt_labels (Tensor) –

  Ground-truth class labels (\(1\)-based foreground indices), shape \((G,)\).

• matched_threshold (float) –

  IoU at or above which an anchor becomes a positive.

• unmatched_threshold (float) –

  IoU below which an anchor becomes background.

• match_height (bool, default: False ) –

  Match by the rotated 3D IoU when True, otherwise by the bird's-eye IoU of the boxes snapped to their nearest axis-aligned orientation (boxes_iou_nearest_bev).

• gt_batch (Optional[Tensor], default: None ) –

  Scene index of every ground-truth box, shape \((G,)\), to match a whole batch at once.

• batch_size (Optional[int], default: None ) –

  Number of scenes \(B\) when gt_batch is given.

Returns:

• AnchorTargets –

  A TypedDict with cls_labels \((A,)\) (\(-1\) ignore, \(0\) background, \(\ge 1\) foreground class) and

• AnchorTargets –

  box_reg_targets \((A, 7 + C)\) (residual encodings, zero for non-positive anchors); \((B, A)\) and

• AnchorTargets –

  \((B, A, 7 + C)\) with gt_batch.

>>> anchors = torch.tensor([[0.0, 0.0, 0.0, 4.0, 2.0, 1.5, 0.0], [20.0, 0.0, 0.0, 4.0, 2.0, 1.5, 0.0]])
>>> gt_boxes = torch.tensor([[0.0, 0.0, 0.0, 4.0, 2.0, 1.5, 0.0]])
>>> gt_labels = torch.tensor([1])
>>> out = assign_anchor_targets(anchors, gt_boxes, gt_labels, matched_threshold=0.6, unmatched_threshold=0.45)
>>> out["cls_labels"].tolist()
[1, 0]