Python mmdet.core.bbox2roi() Examples
The following are 30
code examples of mmdet.core.bbox2roi().
You can vote up the ones you like or vote down the ones you don't like,
and go to the original project or source file by following the links above each example.
You may also want to check out all available functions/classes of the module
mmdet.core
, or try the search function
.
.
Example #1
| Source File: grid_rcnn.py From Feature-Selective-Anchor-Free-Module-for-Single-Shot-Object-Detection with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #2
| Source File: grid_rcnn.py From ttfnet with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #3
| Source File: dynamic_roi_head.py From mmdetection with Apache License 2.0 | 6 votes |
|
def _bbox_forward_train(self, x, sampling_results, gt_bboxes, gt_labels,
img_metas):
num_imgs = len(img_metas)
rois = bbox2roi([res.bboxes for res in sampling_results])
bbox_results = self._bbox_forward(x, rois)
bbox_targets = self.bbox_head.get_targets(sampling_results, gt_bboxes,
gt_labels, self.train_cfg)
# record the `beta_topk`-th smallest target
# `bbox_targets[2]` and `bbox_targets[3]` stand for bbox_targets
# and bbox_weights, respectively
pos_inds = bbox_targets[3][:, 0].nonzero().squeeze(1)
num_pos = len(pos_inds)
cur_target = bbox_targets[2][pos_inds, :2].abs().mean(dim=1)
beta_topk = min(self.train_cfg.dynamic_rcnn.beta_topk * num_imgs,
num_pos)
cur_target = torch.kthvalue(cur_target, beta_topk)[0].item()
self.beta_history.append(cur_target)
loss_bbox = self.bbox_head.loss(bbox_results['cls_score'],
bbox_results['bbox_pred'], rois,
*bbox_targets)
bbox_results.update(loss_bbox=loss_bbox)
return bbox_results
Example #4
| Source File: grid_roi_head.py From mmdetection with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, x, proposals):
"""Dummy forward function."""
# bbox head
outs = ()
rois = bbox2roi([proposals])
if self.with_bbox:
bbox_results = self._bbox_forward(x, rois)
outs = outs + (bbox_results['cls_score'],
bbox_results['bbox_pred'])
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
outs = outs + (grid_pred, )
# mask head
if self.with_mask:
mask_rois = rois[:100]
mask_results = self._mask_forward(x, mask_rois)
outs = outs + (mask_results['mask_pred'], )
return outs
Example #5
| Source File: double_head_rcnn.py From ttfnet with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #6
| Source File: double_head_rcnn.py From Feature-Selective-Anchor-Free-Module-for-Single-Shot-Object-Detection with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #7
| Source File: point_rend_roi_head.py From mmdetection with Apache License 2.0 | 6 votes |
|
def _mask_point_forward_train(self, x, sampling_results, mask_pred,
gt_masks, img_metas):
"""Run forward function and calculate loss for point head in
training."""
pos_labels = torch.cat([res.pos_gt_labels for res in sampling_results])
rel_roi_points = self.point_head.get_roi_rel_points_train(
mask_pred, pos_labels, cfg=self.train_cfg)
rois = bbox2roi([res.pos_bboxes for res in sampling_results])
fine_grained_point_feats = self._get_fine_grained_point_feats(
x, rois, rel_roi_points, img_metas)
coarse_point_feats = point_sample(mask_pred, rel_roi_points)
mask_point_pred = self.point_head(fine_grained_point_feats,
coarse_point_feats)
mask_point_target = self.point_head.get_targets(
rois, rel_roi_points, sampling_results, gt_masks, self.train_cfg)
loss_mask_point = self.point_head.loss(mask_point_pred,
mask_point_target, pos_labels)
return loss_mask_point
Example #8
| Source File: double_head_rcnn.py From Cascade-RPN with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #9
| Source File: grid_rcnn.py From PolarMask with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #10
| Source File: double_head_rcnn.py From PolarMask with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #11
| Source File: grid_rcnn.py From kaggle-kuzushiji-recognition with MIT License | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #12
| Source File: grid_rcnn.py From Cascade-RPN with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #13
| Source File: double_head_rcnn.py From FoveaBox with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #14
| Source File: double_head_rcnn.py From kaggle-kuzushiji-recognition with MIT License | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #15
| Source File: grid_rcnn.py From RDSNet with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #16
| Source File: grid_rcnn.py From FoveaBox with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #17
| Source File: double_head_rcnn.py From RDSNet with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #18
| Source File: grid_rcnn.py From IoU-Uniform-R-CNN with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
# grid head
grid_rois = rois[:100]
grid_feats = self.grid_roi_extractor(
x[:self.grid_roi_extractor.num_inputs], grid_rois)
if self.with_shared_head:
grid_feats = self.shared_head(grid_feats)
grid_pred = self.grid_head(grid_feats)
return rpn_outs, cls_score, bbox_pred, grid_pred
Example #19
| Source File: double_head_rcnn.py From IoU-Uniform-R-CNN with Apache License 2.0 | 6 votes |
|
def forward_dummy(self, img):
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
outs += (cls_score, bbox_pred)
return outs
Example #20
| Source File: grid_rcnn.py From CenterNet with Apache License 2.0 | 5 votes |
|
def simple_test(self, img, img_meta, proposals=None, rescale=False):
"""Test without augmentation."""
assert self.with_bbox, "Bbox head must be implemented."
x = self.extract_feat(img)
proposal_list = self.simple_test_rpn(
x, img_meta, self.test_cfg.rpn) if proposals is None else proposals
det_bboxes, det_labels = self.simple_test_bboxes(
x, img_meta, proposal_list, self.test_cfg.rcnn, rescale=False)
# pack rois into bboxes
grid_rois = bbox2roi([det_bboxes[:, :4]])
grid_feats = self.grid_roi_extractor(
x[:len(self.grid_roi_extractor.featmap_strides)], grid_rois)
if grid_rois.shape[0] != 0:
self.grid_head.test_mode = True
grid_pred = self.grid_head(grid_feats)
det_bboxes = self.grid_head.get_bboxes(det_bboxes,
grid_pred['fused'],
img_meta)
if rescale:
det_bboxes[:, :4] /= img_meta[0]['scale_factor']
else:
det_bboxes = torch.Tensor([])
bbox_results = bbox2result(det_bboxes, det_labels,
self.bbox_head.num_classes)
return bbox_results
Example #21
| Source File: point_rend_roi_head.py From mmdetection with Apache License 2.0 | 5 votes |
|
def simple_test_mask(self,
x,
img_metas,
det_bboxes,
det_labels,
rescale=False):
"""Obtain mask prediction without augmentation."""
# image shape of the first image in the batch (only one)
ori_shape = img_metas[0]['ori_shape']
scale_factor = img_metas[0]['scale_factor']
if det_bboxes.shape[0] == 0:
segm_result = [[] for _ in range(self.mask_head.num_classes)]
else:
# if det_bboxes is rescaled to the original image size, we need to
# rescale it back to the testing scale to obtain RoIs.
if rescale and not isinstance(scale_factor, float):
scale_factor = det_bboxes.new_tensor(scale_factor)
_bboxes = (
det_bboxes[:, :4] * scale_factor if rescale else det_bboxes)
mask_rois = bbox2roi([_bboxes])
mask_results = self._mask_forward(x, mask_rois)
mask_results['mask_pred'] = self._mask_point_forward_test(
x, mask_rois, det_labels, mask_results['mask_pred'], img_metas)
segm_result = self.mask_head.get_seg_masks(
mask_results['mask_pred'], _bboxes, det_labels, self.test_cfg,
ori_shape, scale_factor, rescale)
return segm_result
Example #22
| Source File: mask_scoring_rcnn.py From Cascade-RPN with Apache License 2.0 | 5 votes |
|
def simple_test_mask(self,
x,
img_meta,
det_bboxes,
det_labels,
rescale=False):
# image shape of the first image in the batch (only one)
ori_shape = img_meta[0]['ori_shape']
scale_factor = img_meta[0]['scale_factor']
if det_bboxes.shape[0] == 0:
segm_result = [[] for _ in range(self.mask_head.num_classes - 1)]
mask_scores = [[] for _ in range(self.mask_head.num_classes - 1)]
else:
# if det_bboxes is rescaled to the original image size, we need to
# rescale it back to the testing scale to obtain RoIs.
_bboxes = (
det_bboxes[:, :4] * scale_factor if rescale else det_bboxes)
mask_rois = bbox2roi([_bboxes])
mask_feats = self.mask_roi_extractor(
x[:len(self.mask_roi_extractor.featmap_strides)], mask_rois)
if self.with_shared_head:
mask_feats = self.shared_head(mask_feats)
mask_pred = self.mask_head(mask_feats)
segm_result = self.mask_head.get_seg_masks(mask_pred, _bboxes,
det_labels,
self.test_cfg.rcnn,
ori_shape, scale_factor,
rescale)
# get mask scores with mask iou head
mask_iou_pred = self.mask_iou_head(
mask_feats, mask_pred[range(det_labels.size(0)),
det_labels + 1])
mask_scores = self.mask_iou_head.get_mask_scores(
mask_iou_pred, det_bboxes, det_labels)
return segm_result, mask_scores
Example #23
| Source File: double_head_rcnn.py From Cascade-RPN with Apache License 2.0 | 5 votes |
|
def simple_test_bboxes(self,
x,
img_meta,
proposals,
rcnn_test_cfg,
rescale=False):
"""Test only det bboxes without augmentation."""
rois = bbox2roi(proposals)
bbox_cls_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
bbox_reg_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs],
rois,
roi_scale_factor=self.reg_roi_scale_factor)
if self.with_shared_head:
bbox_cls_feats = self.shared_head(bbox_cls_feats)
bbox_reg_feats = self.shared_head(bbox_reg_feats)
cls_score, bbox_pred = self.bbox_head(bbox_cls_feats, bbox_reg_feats)
img_shape = img_meta[0]['img_shape']
scale_factor = img_meta[0]['scale_factor']
det_bboxes, det_labels = self.bbox_head.get_det_bboxes(
rois,
cls_score,
bbox_pred,
img_shape,
scale_factor,
rescale=rescale,
cfg=rcnn_test_cfg)
return det_bboxes, det_labels
Example #24
| Source File: mask_scoring_rcnn.py From FoveaBox with Apache License 2.0 | 5 votes |
|
def simple_test_mask(self,
x,
img_meta,
det_bboxes,
det_labels,
rescale=False):
# image shape of the first image in the batch (only one)
ori_shape = img_meta[0]['ori_shape']
scale_factor = img_meta[0]['scale_factor']
if det_bboxes.shape[0] == 0:
segm_result = [[] for _ in range(self.mask_head.num_classes - 1)]
mask_scores = [[] for _ in range(self.mask_head.num_classes - 1)]
else:
# if det_bboxes is rescaled to the original image size, we need to
# rescale it back to the testing scale to obtain RoIs.
_bboxes = (
det_bboxes[:, :4] * scale_factor if rescale else det_bboxes)
mask_rois = bbox2roi([_bboxes])
mask_feats = self.mask_roi_extractor(
x[:len(self.mask_roi_extractor.featmap_strides)], mask_rois)
if self.with_shared_head:
mask_feats = self.shared_head(mask_feats)
mask_pred = self.mask_head(mask_feats)
segm_result = self.mask_head.get_seg_masks(mask_pred, _bboxes,
det_labels,
self.test_cfg.rcnn,
ori_shape, scale_factor,
rescale)
# get mask scores with mask iou head
mask_iou_pred = self.mask_iou_head(
mask_feats,
mask_pred[range(det_labels.size(0)), det_labels + 1])
mask_scores = self.mask_iou_head.get_mask_scores(
mask_iou_pred, det_bboxes, det_labels)
return segm_result, mask_scores
Example #25
| Source File: grid_rcnn.py From FoveaBox with Apache License 2.0 | 5 votes |
|
def simple_test(self, img, img_meta, proposals=None, rescale=False):
"""Test without augmentation."""
assert self.with_bbox, "Bbox head must be implemented."
x = self.extract_feat(img)
proposal_list = self.simple_test_rpn(
x, img_meta, self.test_cfg.rpn) if proposals is None else proposals
det_bboxes, det_labels = self.simple_test_bboxes(
x, img_meta, proposal_list, self.test_cfg.rcnn, rescale=False)
# pack rois into bboxes
grid_rois = bbox2roi([det_bboxes[:, :4]])
grid_feats = self.grid_roi_extractor(
x[:len(self.grid_roi_extractor.featmap_strides)], grid_rois)
if grid_rois.shape[0] != 0:
self.grid_head.test_mode = True
grid_pred = self.grid_head(grid_feats)
det_bboxes = self.grid_head.get_bboxes(det_bboxes,
grid_pred['fused'],
img_meta)
if rescale:
det_bboxes[:, :4] /= img_meta[0]['scale_factor']
else:
det_bboxes = torch.Tensor([])
bbox_results = bbox2result(det_bboxes, det_labels,
self.bbox_head.num_classes)
return bbox_results
Example #26
| Source File: two_stage.py From ttfnet with Apache License 2.0 | 5 votes |
|
def forward_dummy(self, img):
"""Used for computing network flops.
See `mmedetection/tools/get_flops.py`
"""
outs = ()
# backbone
x = self.extract_feat(img)
# rpn
if self.with_rpn:
rpn_outs = self.rpn_head(x)
outs = outs + (rpn_outs, )
proposals = torch.randn(1000, 4).cuda()
# bbox head
rois = bbox2roi([proposals])
if self.with_bbox:
bbox_feats = self.bbox_roi_extractor(
x[:self.bbox_roi_extractor.num_inputs], rois)
if self.with_shared_head:
bbox_feats = self.shared_head(bbox_feats)
cls_score, bbox_pred = self.bbox_head(bbox_feats)
outs = outs + (cls_score, bbox_pred)
# mask head
if self.with_mask:
mask_rois = rois[:100]
mask_feats = self.mask_roi_extractor(
x[:self.mask_roi_extractor.num_inputs], mask_rois)
if self.with_shared_head:
mask_feats = self.shared_head(mask_feats)
mask_pred = self.mask_head(mask_feats)
outs = outs + (mask_pred, )
return outs
Example #27
| Source File: mask_scoring_rcnn.py From Libra_R-CNN with Apache License 2.0 | 5 votes |
|
def simple_test_mask(self,
x,
img_meta,
det_bboxes,
det_labels,
rescale=False):
# image shape of the first image in the batch (only one)
ori_shape = img_meta[0]['ori_shape']
scale_factor = img_meta[0]['scale_factor']
if det_bboxes.shape[0] == 0:
segm_result = [[] for _ in range(self.mask_head.num_classes - 1)]
mask_scores = [[] for _ in range(self.mask_head.num_classes - 1)]
else:
# if det_bboxes is rescaled to the original image size, we need to
# rescale it back to the testing scale to obtain RoIs.
_bboxes = (
det_bboxes[:, :4] * scale_factor if rescale else det_bboxes)
mask_rois = bbox2roi([_bboxes])
mask_feats = self.mask_roi_extractor(
x[:len(self.mask_roi_extractor.featmap_strides)], mask_rois)
if self.with_shared_head:
mask_feats = self.shared_head(mask_feats)
mask_pred = self.mask_head(mask_feats)
segm_result = self.mask_head.get_seg_masks(mask_pred, _bboxes,
det_labels,
self.test_cfg.rcnn,
ori_shape, scale_factor,
rescale)
# get mask scores with mask iou head
mask_iou_pred = self.mask_iou_head(
mask_feats,
mask_pred[range(det_labels.size(0)), det_labels + 1])
mask_scores = self.mask_iou_head.get_mask_scores(
mask_iou_pred, det_bboxes, det_labels)
return segm_result, mask_scores
Example #28
| Source File: grid_rcnn.py From Libra_R-CNN with Apache License 2.0 | 5 votes |
|
def simple_test(self, img, img_meta, proposals=None, rescale=False):
"""Test without augmentation."""
assert self.with_bbox, "Bbox head must be implemented."
x = self.extract_feat(img)
proposal_list = self.simple_test_rpn(
x, img_meta, self.test_cfg.rpn) if proposals is None else proposals
det_bboxes, det_labels = self.simple_test_bboxes(
x, img_meta, proposal_list, self.test_cfg.rcnn, rescale=False)
# pack rois into bboxes
grid_rois = bbox2roi([det_bboxes[:, :4]])
grid_feats = self.grid_roi_extractor(
x[:len(self.grid_roi_extractor.featmap_strides)], grid_rois)
if grid_rois.shape[0] != 0:
self.grid_head.test_mode = True
grid_pred = self.grid_head(grid_feats)
det_bboxes = self.grid_head.get_bboxes(det_bboxes,
grid_pred['fused'],
img_meta)
if rescale:
det_bboxes[:, :4] /= img_meta[0]['scale_factor']
else:
det_bboxes = torch.Tensor([])
bbox_results = bbox2result(det_bboxes, det_labels,
self.bbox_head.num_classes)
return bbox_results
Example #29
| Source File: hkrm_rcnn.py From Reasoning-RCNN with Apache License 2.0 | 5 votes |
|
def simple_test_bboxes_hkrm(self,
x,
img_meta,
proposals,
rcnn_test_cfg,
rescale=False,
use_hkrm=False):
"""Test only det bboxes without augmentation."""
rois = bbox2roi(proposals)
roi_feats = self.bbox_roi_extractor(
x[:len(self.bbox_roi_extractor.featmap_strides)], rois)
if self.with_upper_neck:
roi_feats = self.upper_neck(roi_feats)
if use_hkrm:
# Get grometric feature of rois
geom_f = []
for img_i, img_shape in enumerate(img_meta):
h, w, _ = img_meta[img_i]['pad_shape']
tmp_geo = rois[rois[:, 0] == img_i, 1:] / torch.Tensor([h, w, h, w]).cuda()
tmp_geo = torch.cat((tmp_geo, proposals[img_i][:, 4].unsqueeze(1)), dim=-1)
geom_f.append(tmp_geo)
geom_f = torch.stack(geom_f, 0)
assert len(geom_f.size()) == 3
cls_score, bbox_pred, A_pred = self.bbox_head_hkrm(roi_feats, geom_f, len(img_meta))
else:
cls_score, bbox_pred = self.bbox_head_hkrm(roi_feats)
img_shape = img_meta[0]['img_shape']
scale_factor = img_meta[0]['scale_factor']
det_bboxes, det_labels = self.bbox_head_hkrm.get_det_bboxes(
rois,
cls_score,
bbox_pred,
img_shape,
scale_factor,
rescale=rescale,
cfg=rcnn_test_cfg)
return det_bboxes, det_labels
Example #30
| Source File: mask_scoring_rcnn.py From ttfnet with Apache License 2.0 | 5 votes |
|
def simple_test_mask(self,
x,
img_meta,
det_bboxes,
det_labels,
rescale=False):
# image shape of the first image in the batch (only one)
ori_shape = img_meta[0]['ori_shape']
scale_factor = img_meta[0]['scale_factor']
if det_bboxes.shape[0] == 0:
segm_result = [[] for _ in range(self.mask_head.num_classes - 1)]
mask_scores = [[] for _ in range(self.mask_head.num_classes - 1)]
else:
# if det_bboxes is rescaled to the original image size, we need to
# rescale it back to the testing scale to obtain RoIs.
_bboxes = (
det_bboxes[:, :4] * scale_factor if rescale else det_bboxes)
mask_rois = bbox2roi([_bboxes])
mask_feats = self.mask_roi_extractor(
x[:len(self.mask_roi_extractor.featmap_strides)], mask_rois)
if self.with_shared_head:
mask_feats = self.shared_head(mask_feats)
mask_pred = self.mask_head(mask_feats)
segm_result = self.mask_head.get_seg_masks(mask_pred, _bboxes,
det_labels,
self.test_cfg.rcnn,
ori_shape, scale_factor,
rescale)
# get mask scores with mask iou head
mask_iou_pred = self.mask_iou_head(
mask_feats,
mask_pred[range(det_labels.size(0)), det_labels + 1])
mask_scores = self.mask_iou_head.get_mask_scores(
mask_iou_pred, det_bboxes, det_labels)
return segm_result, mask_scores
