def _bboxes_nms(self, bboxes, labels, cfg):
if labels.numel() == 0:
return bboxes, labels
if 'nms_cfg' in cfg:
warning.warn('nms_cfg in test_cfg will be deprecated. '
'Please rename it as nms')
if 'nms' not in cfg:
cfg.nms = cfg.nms_cfg
out_bboxes, keep = batched_nms(bboxes[:, :4], bboxes[:, -1], labels,
cfg.nms)
out_labels = labels[keep]
if len(out_bboxes) > 0:
# use `sort` to replace with `argsort` here
_, idx = torch.sort(out_bboxes[:, -1], descending=True)
max_per_img = out_bboxes.new_tensor(cfg.max_per_img).to(torch.long)
nms_after = max_per_img
if torch.onnx.is_in_onnx_export():
# Always keep topk op for dynamic input in onnx
from mmdet.core.export import get_k_for_topk
nms_after = get_k_for_topk(max_per_img, out_bboxes.shape[0])
idx = idx[:nms_after]
out_bboxes = out_bboxes[idx]
out_labels = out_labels[idx]
return out_bboxes, out_labels
def __init__(self, session=None, de=None):
if session is None:
session = database.create_session()
self.session = session
if de is not None:
warning.warn(
"DatabaseExtension has been deprecated is no longer supported")
self.experiments = querying.get_experiments(session=self.session,
all=True)
self.keywords = sorted(querying.get_keywords(self.session),
key=str.casefold)
self.variables = querying.get_variables(self.session, inferred=True)
self._make_widgets()
# Call super init and pass widgets as children
super().__init__(children=[
self.header, self.selectors, self.expt_info, self.expt_explorer
])
# Show the experiment information: important for only one experiment, as
# events will not trigger this otherwise
self._show_experiment_information(self.expt_selector.value)
self._set_handlers()
def export_2d_annotation(root_path, info_path, version):
"""Export 2d annotation from the info file and raw data.
Args:
root_path (str): Root path of the raw data.
info_path (str): Path of the info file.
version (str): Dataset version.
"""
warning.warn('DeprecationWarning: 2D annotations are not used on the '
'Lyft dataset. The function export_2d_annotation will be '
'deprecated.')
# get bbox annotations for camera
camera_types = [
'CAM_FRONT',
'CAM_FRONT_RIGHT',
'CAM_FRONT_LEFT',
'CAM_BACK',
'CAM_BACK_LEFT',
'CAM_BACK_RIGHT',
]
lyft_infos = mmcv.load(info_path)['infos']
lyft = Lyft(data_path=osp.join(root_path, version),
json_path=osp.join(root_path, version, version),
verbose=True)
# info_2d_list = []
cat2Ids = [
dict(id=lyft_categories.index(cat_name), name=cat_name)
for cat_name in lyft_categories
]
coco_ann_id = 0
coco_2d_dict = dict(annotations=[], images=[], categories=cat2Ids)
for info in mmcv.track_iter_progress(lyft_infos):
for cam in camera_types:
cam_info = info['cams'][cam]
coco_infos = get_2d_boxes(lyft,
cam_info['sample_data_token'],
visibilities=['', '1', '2', '3', '4'])
(height, width, _) = mmcv.imread(cam_info['data_path']).shape
coco_2d_dict['images'].append(
dict(file_name=cam_info['data_path'],
id=cam_info['sample_data_token'],
width=width,
height=height))
for coco_info in coco_infos:
if coco_info is None:
continue
# add an empty key for coco format
coco_info['segmentation'] = []
coco_info['id'] = coco_ann_id
coco_2d_dict['annotations'].append(coco_info)
coco_ann_id += 1
mmcv.dump(coco_2d_dict, f'{info_path[:-4]}.coco.json')
def _bboxes_nms(self, bboxes, labels, cfg):
if 'nms_cfg' in cfg:
warning.warn('nms_cfg in test_cfg will be deprecated. '
'Please rename it as nms')
if 'nms' not in cfg:
cfg.nms = cfg.nms_cfg
if labels.numel() > 0:
max_num = cfg.max_per_img
bboxes, keep = batched_nms(bboxes[:, :4], bboxes[:,
-1].contiguous(),
labels, cfg.nms)
if max_num > 0:
bboxes = bboxes[:max_num]
labels = labels[keep][:max_num]
return bboxes, labels
def mono_cam_box2vis(cam_box):
"""This is a post-processing function on the bboxes from Mono-3D task. If
we want to perform projection visualization, we need to:
1. rotate the box along x-axis for np.pi / 2 (roll)
2. change orientation from local yaw to global yaw
3. convert yaw by (np.pi / 2 - yaw)
After applying this function, we can project and draw it on 2D images.
Args:
cam_box (:obj:`CameraInstance3DBoxes`): 3D bbox in camera coordinate \
system before conversion. Could be gt bbox loaded from dataset or \
network prediction output.
Returns:
:obj:`CameraInstance3DBoxes`: Box after conversion.
"""
warning.warn('DeprecationWarning: The hack of yaw and dimension in the '
'monocular 3D detection on nuScenes has been removed. The '
'function mono_cam_box2vis will be deprecated.')
from . import CameraInstance3DBoxes
assert isinstance(cam_box, CameraInstance3DBoxes), \
'input bbox should be CameraInstance3DBoxes!'
loc = cam_box.gravity_center
dim = cam_box.dims
yaw = cam_box.yaw
feats = cam_box.tensor[:, 7:]
# rotate along x-axis for np.pi / 2
# see also here: https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/nuscenes_mono_dataset.py#L557 # noqa
dim[:, [1, 2]] = dim[:, [2, 1]]
# change local yaw to global yaw for visualization
# refer to https://github.com/open-mmlab/mmdetection3d/blob/master/mmdet3d/datasets/nuscenes_mono_dataset.py#L164-L166 # noqa
yaw += torch.atan2(loc[:, 0], loc[:, 2])
# convert yaw by (-yaw - np.pi / 2)
# this is because mono 3D box class such as `NuScenesBox` has different
# definition of rotation with our `CameraInstance3DBoxes`
yaw = -yaw - np.pi / 2
cam_box = torch.cat([loc, dim, yaw[:, None], feats], dim=1)
cam_box = CameraInstance3DBoxes(cam_box,
box_dim=cam_box.shape[-1],
origin=(0.5, 0.5, 0.5))
return cam_box
def _bboxes_nms(self, bboxes, labels, cfg):
if labels.numel() == 0:
return bboxes, labels
if 'nms_cfg' in cfg:
warning.warn('nms_cfg in test_cfg will be deprecated. '
'Please rename it as nms')
if 'nms' not in cfg:
cfg.nms = cfg.nms_cfg
out_bboxes, keep = batched_nms(bboxes[:, :4], bboxes[:, -1], labels,
cfg.nms)
out_labels = labels[keep]
if len(out_bboxes) > 0:
idx = torch.argsort(out_bboxes[:, -1], descending=True)
idx = idx[:cfg.max_per_img]
out_bboxes = out_bboxes[idx]
out_labels = out_labels[idx]
return out_bboxes, out_labels