Spaces:
Sleeping
Sleeping
File size: 3,438 Bytes
749745d |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 |
# Copyright (c) Facebook, Inc. and its affiliates. All Rights Reserved.
import torch
from torch import nn
from maskrcnn_benchmark.structures.bounding_box import BoxList
from .roi_mask_feature_extractors import make_roi_mask_feature_extractor
from .roi_mask_predictors import make_roi_mask_predictor
from .inference import make_roi_mask_post_processor
from .loss import make_roi_mask_loss_evaluator
def keep_only_positive_boxes(boxes):
"""
Given a set of BoxList containing the `labels` field,
return a set of BoxList for which `labels > 0`.
Arguments:
boxes (list of BoxList)
"""
assert isinstance(boxes, (list, tuple))
assert isinstance(boxes[0], BoxList)
assert boxes[0].has_field("labels")
positive_boxes = []
positive_inds = []
num_boxes = 0
for boxes_per_image in boxes:
labels = boxes_per_image.get_field("labels")
inds_mask = labels > 0
inds = inds_mask.nonzero().squeeze(1)
positive_boxes.append(boxes_per_image[inds])
positive_inds.append(inds_mask)
return positive_boxes, positive_inds
class ROIMaskHead(torch.nn.Module):
def __init__(self, cfg):
super(ROIMaskHead, self).__init__()
self.cfg = cfg.clone()
self.feature_extractor = make_roi_mask_feature_extractor(cfg)
self.predictor = make_roi_mask_predictor(cfg)
self.post_processor = make_roi_mask_post_processor(cfg)
self.loss_evaluator = make_roi_mask_loss_evaluator(cfg)
def forward(self, features, proposals, targets=None, language_dict_features=None, positive_map_label_to_token=None):
"""
Arguments:
features (list[Tensor]): feature-maps from possibly several levels
proposals (list[BoxList]): proposal boxes
targets (list[BoxList], optional): the ground-truth targets.
language_dict_features: language features: hidden, embedding, mask, ...
Returns:
x (Tensor): the result of the feature extractor
proposals (list[BoxList]): during training, the original proposals
are returned. During testing, the predicted boxlists are returned
with the `mask` field set
losses (dict[Tensor]): During training, returns the losses for the
head. During testing, returns an empty dict.
"""
if self.training:
# during training, only focus on positive boxes
all_proposals = proposals
proposals, positive_inds = keep_only_positive_boxes(proposals)
if self.training and self.cfg.MODEL.ROI_MASK_HEAD.SHARE_BOX_FEATURE_EXTRACTOR:
x = features
x = x[torch.cat(positive_inds, dim=0)]
else:
x = self.feature_extractor(features, proposals)
if self.cfg.MODEL.ROI_MASK_HEAD.PREDICTOR.startswith("VL"):
mask_logits = self.predictor(x, language_dict_features)
else:
mask_logits = self.predictor(x)
if not self.training:
result = self.post_processor(mask_logits, proposals, positive_map_label_to_token)
return x, result, {}
loss_mask = self.loss_evaluator(proposals, mask_logits, targets)
return x, all_proposals, dict(loss_mask=loss_mask)
def build_roi_mask_head(cfg):
return ROIMaskHead(cfg)
|