def predict(self, example, preds_dict):
t = time.time()
batch_size = example['anchors'].shape[0]
batch_anchors = example["anchors"].view(batch_size, -1, 7)
self._total_inference_count += batch_size
batch_rect = example["rect"]
batch_Trv2c = example["Trv2c"]
batch_P2 = example["P2"]
if "anchors_mask" not in example:
batch_anchors_mask = [None] * batch_size
else:
batch_anchors_mask = example["anchors_mask"].view(batch_size, -1)
batch_imgidx = example['image_idx']
self._total_forward_time += time.time() - t
t = time.time()
batch_box_preds = preds_dict["box_preds"]
batch_cls_preds = preds_dict["cls_preds"]
batch_box_preds = batch_box_preds.view(batch_size, -1,
self._box_coder.code_size)
num_class_with_bg = self._num_class
if not self._encode_background_as_zeros:
num_class_with_bg = self._num_class + 1
batch_cls_preds = batch_cls_preds.view(batch_size, -1,
num_class_with_bg)
batch_box_preds = self._box_coder.decode_torch(batch_box_preds,
batch_anchors)
if self._use_direction_classifier:
batch_dir_preds = preds_dict["dir_cls_preds"]
batch_dir_preds = batch_dir_preds.view(batch_size, -1, 2)
else:
batch_dir_preds = [None] * batch_size
predictions_dicts = []
for box_preds, cls_preds, dir_preds, rect, Trv2c, P2, img_idx, a_mask in zip(
batch_box_preds, batch_cls_preds, batch_dir_preds, batch_rect,
batch_Trv2c, batch_P2, batch_imgidx, batch_anchors_mask
):
if a_mask is not None:
box_preds = box_preds[a_mask]
cls_preds = cls_preds[a_mask]
if self._use_direction_classifier:
if a_mask is not None:
dir_preds = dir_preds[a_mask]
# print(dir_preds.shape)
dir_labels = torch.max(dir_preds, dim=-1)[1]
if self._encode_background_as_zeros:
# this don't support softmax
assert self._use_sigmoid_score is True
total_scores = torch.sigmoid(cls_preds)
else:
# encode background as first element in one-hot vector
if self._use_sigmoid_score:
total_scores = torch.sigmoid(cls_preds)[..., 1:]
else:
total_scores = F.softmax(cls_preds, dim=-1)[..., 1:]
# Apply NMS in birdeye view
if self._use_rotate_nms:
nms_func = box_torch_ops.rotate_nms
else:
nms_func = box_torch_ops.nms
selected_boxes = None
selected_labels = None
selected_scores = None
selected_dir_labels = None
if self._multiclass_nms:
# curently only support class-agnostic boxes.
boxes_for_nms = box_preds[:, [0, 1, 3, 4, 6]]
if not self._use_rotate_nms:
box_preds_corners = box_torch_ops.center_to_corner_box2d(
boxes_for_nms[:, :2], boxes_for_nms[:, 2:4],
boxes_for_nms[:, 4])
boxes_for_nms = box_torch_ops.corner_to_standup_nd(
box_preds_corners)
boxes_for_mcnms = boxes_for_nms.unsqueeze(1)
selected_per_class = box_torch_ops.multiclass_nms(
nms_func=nms_func,
boxes=boxes_for_mcnms,
scores=total_scores,
num_class=self._num_class,
pre_max_size=self._nms_pre_max_size,
post_max_size=self._nms_post_max_size,
iou_threshold=self._nms_iou_threshold,
score_thresh=self._nms_score_threshold,
)
selected_boxes, selected_labels, selected_scores = [], [], []
selected_dir_labels = []
for i, selected in enumerate(selected_per_class):
if selected is not None:
num_dets = selected.shape[0]
selected_boxes.append(box_preds[selected])
selected_labels.append(
torch.full([num_dets], i, dtype=torch.int64))
if self._use_direction_classifier:
selected_dir_labels.append(dir_labels[selected])
selected_scores.append(total_scores[selected, i])
if len(selected_boxes) > 0:
selected_boxes = torch.cat(selected_boxes, dim=0)
selected_labels = torch.cat(selected_labels, dim=0)
selected_scores = torch.cat(selected_scores, dim=0)
if self._use_direction_classifier:
selected_dir_labels = torch.cat(
selected_dir_labels, dim=0)
else:
selected_boxes = None
selected_labels = None
selected_scores = None
selected_dir_labels = None
else:
# get highest score per prediction, than apply nms
# to remove overlapped box.
if num_class_with_bg == 1:
top_scores = total_scores.squeeze(-1)
top_labels = torch.zeros(
total_scores.shape[0],
device=total_scores.device,
dtype=torch.long)
else:
top_scores, top_labels = torch.max(total_scores, dim=-1)
if self._nms_score_threshold > 0.0:
thresh = torch.tensor(
[self._nms_score_threshold],
device=total_scores.device).type_as(total_scores)
top_scores_keep = (top_scores >= thresh)
top_scores = top_scores.masked_select(top_scores_keep)
if top_scores.shape[0] != 0:
if self._nms_score_threshold > 0.0:
box_preds = box_preds[top_scores_keep]
if self._use_direction_classifier:
dir_labels = dir_labels[top_scores_keep]
top_labels = top_labels[top_scores_keep]
boxes_for_nms = box_preds[:, [0, 1, 3, 4, 6]]
if not self._use_rotate_nms:
box_preds_corners = box_torch_ops.center_to_corner_box2d(
boxes_for_nms[:, :2], boxes_for_nms[:, 2:4],
boxes_for_nms[:, 4])
boxes_for_nms = box_torch_ops.corner_to_standup_nd(
box_preds_corners)
# the nms in 3d detection just remove overlap boxes.
selected = nms_func(
boxes_for_nms,
top_scores,
pre_max_size=self._nms_pre_max_size,
post_max_size=self._nms_post_max_size,
iou_threshold=self._nms_iou_threshold,
)
else:
selected = None
if selected is not None:
selected_boxes = box_preds[selected]
if self._use_direction_classifier:
selected_dir_labels = dir_labels[selected]
selected_labels = top_labels[selected]
selected_scores = top_scores[selected]
# finally generate predictions.
if selected_boxes is not None:
box_preds = selected_boxes
scores = selected_scores
label_preds = selected_labels
if self._use_direction_classifier:
dir_labels = selected_dir_labels
opp_labels = (box_preds[..., -1] > 0) ^ dir_labels.byte()
box_preds[..., -1] += torch.where(
opp_labels,
torch.tensor(np.pi).type_as(box_preds),
torch.tensor(0.0).type_as(box_preds))
# box_preds[..., -1] += (
# ~(dir_labels.byte())).type_as(box_preds) * np.pi
final_box_preds = box_preds
final_scores = scores
final_labels = label_preds
final_box_preds_camera = box_torch_ops.box_lidar_to_camera(
final_box_preds, rect, Trv2c)
locs = final_box_preds_camera[:, :3]
dims = final_box_preds_camera[:, 3:6]
angles = final_box_preds_camera[:, 6]
camera_box_origin = [0.5, 1.0, 0.5]
box_corners = box_torch_ops.center_to_corner_box3d(
locs, dims, angles, camera_box_origin, axis=1)
box_corners_in_image = box_torch_ops.project_to_image(
box_corners, P2)
# box_corners_in_image: [N, 8, 2]
minxy = torch.min(box_corners_in_image, dim=1)[0]
maxxy = torch.max(box_corners_in_image, dim=1)[0]
# minx = torch.min(box_corners_in_image[..., 0], dim=1)[0]
# maxx = torch.max(box_corners_in_image[..., 0], dim=1)[0]
# miny = torch.min(box_corners_in_image[..., 1], dim=1)[0]
# maxy = torch.max(box_corners_in_image[..., 1], dim=1)[0]
# box_2d_preds = torch.stack([minx, miny, maxx, maxy], dim=1)
box_2d_preds = torch.cat([minxy, maxxy], dim=1)
# predictions
predictions_dict = {
"bbox": box_2d_preds,
"box3d_camera": final_box_preds_camera,
"box3d_lidar": final_box_preds,
"scores": final_scores,
"label_preds": label_preds,
"image_idx": img_idx,
}
else:
predictions_dict = {
"bbox": None,
"box3d_camera": None,
"box3d_lidar": None,
"scores": None,
"label_preds": None,
"image_idx": img_idx,
}
predictions_dicts.append(predictions_dict)
self._total_postprocess_time += time.time() - t
return predictions_dicts
def predict(self, example, preds_dict):
"""start with v1.6.0, this function don't contain any kitti-specific code.
Returns:
predict: list of pred_dict.
pred_dict: {
box3d_lidar: [N, 7] 3d box.
scores: [N]
label_preds: [N]
metadata: meta-data which contains dataset-specific information.
for kitti, it contains image idx (label idx),
for nuscenes, sample_token is saved in it.
}
"""
batch_size = example['anchors'].shape[0]
if "metadata" not in example or len(example["metadata"]) == 0:
meta_list = [None] * batch_size
else:
meta_list = example["metadata"]
batch_anchors = example["anchors"].view(batch_size, -1,
example["anchors"].shape[-1])
if "anchors_mask" not in example:
batch_anchors_mask = [None] * batch_size
else:
batch_anchors_mask = example["anchors_mask"].view(batch_size, -1)
t = time.time()
batch_box_preds = preds_dict["box_preds"]
batch_cls_preds = preds_dict["cls_preds"]
batch_box_preds = batch_box_preds.view(batch_size, -1,
self._box_coder.code_size)
num_class_with_bg = self._num_class
if not self._encode_background_as_zeros:
num_class_with_bg = self._num_class + 1
batch_cls_preds = batch_cls_preds.view(batch_size, -1,
num_class_with_bg)
batch_box_preds = self._box_coder.decode_torch(batch_box_preds,
batch_anchors)
if self._use_direction_classifier:
batch_dir_preds = preds_dict["dir_cls_preds"]
batch_dir_preds = batch_dir_preds.view(batch_size, -1,
self._num_direction_bins)
else:
batch_dir_preds = [None] * batch_size
predictions_dicts = []
post_center_range = None
if len(self._post_center_range) > 0:
post_center_range = torch.tensor(
self._post_center_range,
dtype=batch_box_preds.dtype,
device=batch_box_preds.device).float()
for box_preds, cls_preds, dir_preds, a_mask, meta in zip(
batch_box_preds, batch_cls_preds, batch_dir_preds,
batch_anchors_mask, meta_list):
if a_mask is not None:
box_preds = box_preds[a_mask]
cls_preds = cls_preds[a_mask]
box_preds = box_preds.float()
cls_preds = cls_preds.float()
if self._use_direction_classifier:
if a_mask is not None:
dir_preds = dir_preds[a_mask]
dir_labels = torch.max(dir_preds, dim=-1)[1]
if self._encode_background_as_zeros:
# this don't support softmax
assert self._use_sigmoid_score is True
total_scores = torch.sigmoid(cls_preds)
else:
# encode background as first element in one-hot vector
if self._use_sigmoid_score:
total_scores = torch.sigmoid(cls_preds)[..., 1:]
else:
total_scores = F.softmax(cls_preds, dim=-1)[..., 1:]
# Apply NMS in birdeye view
if self._use_rotate_nms:
nms_func = box_torch_ops.rotate_nms
else:
nms_func = box_torch_ops.nms
feature_map_size_prod = batch_box_preds.shape[
1] // self.target_assigner.num_anchors_per_location
if self._multiclass_nms:
assert self._encode_background_as_zeros is True
boxes_for_nms = box_preds[:, [0, 1, 3, 4, 6]]
if not self._use_rotate_nms:
box_preds_corners = box_torch_ops.center_to_corner_box2d(
boxes_for_nms[:, :2], boxes_for_nms[:, 2:4],
boxes_for_nms[:, 4])
boxes_for_nms = box_torch_ops.corner_to_standup_nd(
box_preds_corners)
selected_boxes, selected_labels, selected_scores = [], [], []
selected_dir_labels = []
scores = total_scores
boxes = boxes_for_nms
selected_per_class = []
score_threshs = self._nms_score_thresholds
pre_max_sizes = self._nms_pre_max_sizes
post_max_sizes = self._nms_post_max_sizes
iou_thresholds = self._nms_iou_thresholds
for class_idx, score_thresh, pre_ms, post_ms, iou_th in zip(
range(self._num_class), score_threshs, pre_max_sizes,
post_max_sizes, iou_thresholds):
if self._nms_class_agnostic:
class_scores = total_scores.view(
feature_map_size_prod, -1,
self._num_class)[..., class_idx]
class_scores = class_scores.contiguous().view(-1)
class_boxes_nms = boxes.view(-1,
boxes_for_nms.shape[-1])
class_boxes = box_preds
class_dir_labels = dir_labels
else:
anchors_range = self.target_assigner.anchors_range(
class_idx)
class_scores = total_scores.view(
-1,
self._num_class)[anchors_range[0]:anchors_range[1],
class_idx]
class_boxes_nms = boxes.view(
-1, boxes_for_nms.shape[-1])[
anchors_range[0]:anchors_range[1], :]
class_scores = class_scores.contiguous().view(-1)
class_boxes_nms = class_boxes_nms.contiguous().view(
-1, boxes_for_nms.shape[-1])
class_boxes = box_preds.view(-1, box_preds.shape[-1])[
anchors_range[0]:anchors_range[1], :]
class_boxes = class_boxes.contiguous().view(
-1, box_preds.shape[-1])
if self._use_direction_classifier:
class_dir_labels = dir_labels.view(
-1)[anchors_range[0]:anchors_range[1]]
class_dir_labels = class_dir_labels.contiguous(
).view(-1)
if score_thresh > 0.0:
class_scores_keep = class_scores >= score_thresh
if class_scores_keep.shape[0] == 0:
selected_per_class.append(None)
continue
class_scores = class_scores[class_scores_keep]
if class_scores.shape[0] != 0:
if score_thresh > 0.0:
class_boxes_nms = class_boxes_nms[
class_scores_keep]
class_boxes = class_boxes[class_scores_keep]
class_dir_labels = class_dir_labels[
class_scores_keep]
keep = nms_func(class_boxes_nms, class_scores, pre_ms,
post_ms, iou_th)
if keep.shape[0] != 0:
selected_per_class.append(keep)
else:
selected_per_class.append(None)
else:
selected_per_class.append(None)
selected = selected_per_class[-1]
if selected is not None:
selected_boxes.append(class_boxes[selected])
selected_labels.append(
torch.full([class_boxes[selected].shape[0]],
class_idx,
dtype=torch.int64,
device=box_preds.device))
if self._use_direction_classifier:
selected_dir_labels.append(
class_dir_labels[selected])
selected_scores.append(class_scores[selected])
selected_boxes = torch.cat(selected_boxes, dim=0)
selected_labels = torch.cat(selected_labels, dim=0)
selected_scores = torch.cat(selected_scores, dim=0)
if self._use_direction_classifier:
selected_dir_labels = torch.cat(selected_dir_labels, dim=0)
else:
# get highest score per prediction, than apply nms
# to remove overlapped box.
if num_class_with_bg == 1:
top_scores = total_scores.squeeze(-1)
top_labels = torch.zeros(total_scores.shape[0],
device=total_scores.device,
dtype=torch.long)
else:
top_scores, top_labels = torch.max(total_scores, dim=-1)
if self._nms_score_thresholds[0] > 0.0:
top_scores_keep = top_scores >= self._nms_score_thresholds[
0]
top_scores = top_scores.masked_select(top_scores_keep)
if top_scores.shape[0] != 0:
if self._nms_score_thresholds[0] > 0.0:
box_preds = box_preds[top_scores_keep]
if self._use_direction_classifier:
dir_labels = dir_labels[top_scores_keep]
top_labels = top_labels[top_scores_keep]
boxes_for_nms = box_preds[:, [0, 1, 3, 4, 6]]
if not self._use_rotate_nms:
box_preds_corners = box_torch_ops.center_to_corner_box2d(
boxes_for_nms[:, :2], boxes_for_nms[:, 2:4],
boxes_for_nms[:, 4])
boxes_for_nms = box_torch_ops.corner_to_standup_nd(
box_preds_corners)
# the nms in 3d detection just remove overlap boxes.
selected = nms_func(
boxes_for_nms,
top_scores,
pre_max_size=self._nms_pre_max_sizes[0],
post_max_size=self._nms_post_max_sizes[0],
iou_threshold=self._nms_iou_thresholds[0],
)
else:
selected = []
# if selected is not None:
selected_boxes = box_preds[selected]
if self._use_direction_classifier:
selected_dir_labels = dir_labels[selected]
selected_labels = top_labels[selected]
selected_scores = top_scores[selected]
# finally generate predictions.
if selected_boxes.shape[0] != 0:
box_preds = selected_boxes
scores = selected_scores
label_preds = selected_labels
if self._use_direction_classifier:
dir_labels = selected_dir_labels
period = (2 * np.pi / self._num_direction_bins)
dir_rot = box_torch_ops.limit_period(
box_preds[..., 6] - self._dir_offset,
self._dir_limit_offset, period)
box_preds[
...,
6] = dir_rot + self._dir_offset + period * dir_labels.to(
box_preds.dtype)
final_box_preds = box_preds
final_scores = scores
final_labels = label_preds
if post_center_range is not None:
mask = (final_box_preds[:, :3] >=
post_center_range[:3]).all(1)
mask &= (final_box_preds[:, :3] <=
post_center_range[3:]).all(1)
predictions_dict = {
"box3d_lidar": final_box_preds[mask],
"scores": final_scores[mask],
"label_preds": label_preds[mask],
"metadata": meta,
}
else:
predictions_dict = {
"box3d_lidar": final_box_preds,
"scores": final_scores,
"label_preds": label_preds,
"metadata": meta,
}
else:
dtype = batch_box_preds.dtype
device = batch_box_preds.device
predictions_dict = {
"box3d_lidar":
torch.zeros([0, box_preds.shape[-1]],
dtype=dtype,
device=device),
"scores":
torch.zeros([0], dtype=dtype, device=device),
"label_preds":
torch.zeros([0], dtype=top_labels.dtype, device=device),
"metadata":
meta,
}
predictions_dicts.append(predictions_dict)
return predictions_dicts
def predict(self, example, preds_dict):
"""start with v1.6.0, this function don't contain any kitti-specific code.
Returns:
predict: list of pred_dict.
pred_dict: {
box3d_lidar: [N, 7] 3d box.
scores: [N]
label_preds: [N]
metadata: meta-data which contains dataset-specific information.
for kitti, it contains image idx (label idx),
for nuscenes, sample_token is saved in it.
}
"""
batch_size = example['anchors'].shape[0]
if "metadata" not in example or len(example["metadata"]) == 0:
meta_list = [None] * batch_size
else:
meta_list = example["metadata"]
batch_anchors = example["anchors"].view(batch_size, -1, 7)
if "anchors_mask" not in example:
batch_anchors_mask = [None] * batch_size
else:
batch_anchors_mask = example["anchors_mask"].view(batch_size, -1)
t = time.time()
batch_box_preds = preds_dict["box_preds"]
batch_cls_preds = preds_dict["cls_preds"]
batch_box_preds = batch_box_preds.view(batch_size, -1,
self._box_coder.code_size)
num_class_with_bg = self._num_class
if not self._encode_background_as_zeros:
num_class_with_bg = self._num_class + 1
batch_cls_preds = batch_cls_preds.view(batch_size, -1,
num_class_with_bg)
batch_box_preds = self._box_coder.decode_torch(batch_box_preds,
batch_anchors)
if self._use_direction_classifier:
batch_dir_preds = preds_dict["dir_cls_preds"]
batch_dir_preds = batch_dir_preds.view(batch_size, -1, 2)
else:
batch_dir_preds = [None] * batch_size
predictions_dicts = []
post_center_range = None
if len(self._post_center_range) > 0:
post_center_range = torch.tensor(
self._post_center_range,
dtype=batch_box_preds.dtype,
device=batch_box_preds.device).float()
for box_preds, cls_preds, dir_preds, a_mask, meta in zip(
batch_box_preds, batch_cls_preds, batch_dir_preds,
batch_anchors_mask, meta_list):
if a_mask is not None:
box_preds = box_preds[a_mask]
cls_preds = cls_preds[a_mask]
box_preds = box_preds.float()
cls_preds = cls_preds.float()
if self._use_direction_classifier:
if a_mask is not None:
dir_preds = dir_preds[a_mask]
dir_labels = torch.max(dir_preds, dim=-1)[1]
if self._encode_background_as_zeros:
# this don't support softmax
assert self._use_sigmoid_score is True
total_scores = torch.sigmoid(cls_preds)
else:
# encode background as first element in one-hot vector
if self._use_sigmoid_score:
total_scores = torch.sigmoid(cls_preds)[..., 1:]
else:
total_scores = F.softmax(cls_preds, dim=-1)[..., 1:]
# Apply NMS in birdeye view
if self._use_rotate_nms:
nms_func = box_torch_ops.rotate_nms
else:
nms_func = box_torch_ops.nms
if self._multiclass_nms:
# curently only support class-agnostic boxes.
boxes_for_nms = box_preds[:, [0, 1, 3, 4, 6]]
if not self._use_rotate_nms:
box_preds_corners = box_torch_ops.center_to_corner_box2d(
boxes_for_nms[:, :2], boxes_for_nms[:, 2:4],
boxes_for_nms[:, 4])
boxes_for_nms = box_torch_ops.corner_to_standup_nd(
box_preds_corners)
boxes_for_mcnms = boxes_for_nms.unsqueeze(1)
selected_per_class = box_torch_ops.multiclass_nms(
nms_func=nms_func,
boxes=boxes_for_mcnms,
scores=total_scores,
num_class=self._num_class,
pre_max_size=self._nms_pre_max_size,
post_max_size=self._nms_post_max_size,
iou_threshold=self._nms_iou_threshold,
score_thresh=self._nms_score_threshold,
)
selected_boxes, selected_labels, selected_scores = [], [], []
selected_dir_labels = []
for i, selected in enumerate(selected_per_class):
if selected is not None:
num_dets = selected.shape[0]
selected_boxes.append(box_preds[selected])
selected_labels.append(
torch.full([num_dets], i, dtype=torch.int64))
if self._use_direction_classifier:
selected_dir_labels.append(dir_labels[selected])
selected_scores.append(total_scores[selected, i])
selected_boxes = torch.cat(selected_boxes, dim=0)
selected_labels = torch.cat(selected_labels, dim=0)
selected_scores = torch.cat(selected_scores, dim=0)
if self._use_direction_classifier:
selected_dir_labels = torch.cat(selected_dir_labels, dim=0)
else:
# get highest score per prediction, than apply nms
# to remove overlapped box.
if num_class_with_bg == 1:
top_scores = total_scores.squeeze(-1)
top_labels = torch.zeros(total_scores.shape[0],
device=total_scores.device,
dtype=torch.long)
else:
top_scores, top_labels = torch.max(total_scores, dim=-1)
if self._nms_score_threshold > 0.0:
thresh = torch.tensor(
[self._nms_score_threshold],
device=total_scores.device).type_as(total_scores)
top_scores_keep = (top_scores >= thresh)
top_scores = top_scores.masked_select(top_scores_keep)
if top_scores.shape[0] != 0:
if self._nms_score_threshold > 0.0:
box_preds = box_preds[top_scores_keep]
if self._use_direction_classifier:
dir_labels = dir_labels[top_scores_keep]
top_labels = top_labels[top_scores_keep]
boxes_for_nms = box_preds[:, [0, 1, 3, 4, 6]]
if not self._use_rotate_nms:
box_preds_corners = box_torch_ops.center_to_corner_box2d(
boxes_for_nms[:, :2], boxes_for_nms[:, 2:4],
boxes_for_nms[:, 4])
boxes_for_nms = box_torch_ops.corner_to_standup_nd(
box_preds_corners)
# the nms in 3d detection just remove overlap boxes.
selected = nms_func(
boxes_for_nms,
top_scores,
pre_max_size=self._nms_pre_max_size,
post_max_size=self._nms_post_max_size,
iou_threshold=self._nms_iou_threshold,
)
else:
selected = []
# if selected is not None:
selected_boxes = box_preds[selected]
if self._use_direction_classifier:
selected_dir_labels = dir_labels[selected]
selected_labels = top_labels[selected]
selected_scores = top_scores[selected]
# finally generate predictions.
if selected_boxes.shape[0] != 0:
box_preds = selected_boxes
scores = selected_scores
label_preds = selected_labels
if self._use_direction_classifier:
dir_labels = selected_dir_labels
opp_labels = (box_preds[..., -1] > 0) ^ dir_labels.byte()
box_preds[..., -1] += torch.where(
opp_labels,
torch.tensor(np.pi).type_as(box_preds),
torch.tensor(0.0).type_as(box_preds))
final_box_preds = box_preds
final_scores = scores
final_labels = label_preds
if post_center_range is not None:
mask = (final_box_preds[:, :3] >=
post_center_range[:3]).all(1)
mask &= (final_box_preds[:, :3] <=
post_center_range[3:]).all(1)
predictions_dict = {
"box3d_lidar": final_box_preds[mask],
"scores": final_scores[mask],
"label_preds": label_preds[mask],
"metadata": meta,
}
else:
predictions_dict = {
"box3d_lidar": final_box_preds,
"scores": final_scores,
"label_preds": label_preds,
"metadata": meta,
}
else:
dtype = batch_box_preds.dtype
device = batch_box_preds.device
predictions_dict = {
"box3d_lidar":
torch.zeros([0, 7], dtype=dtype, device=device),
"scores":
torch.zeros([0], dtype=dtype, device=device),
"label_preds":
torch.zeros([0], dtype=top_labels.dtype, device=device),
"metadata":
meta,
}
predictions_dicts.append(predictions_dict)
return predictions_dicts
