File size: 6,981 Bytes
3e99b05
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
# coding=utf-8
# Copyright 2022 The IDEA Authors. All rights reserved.
#
# Licensed under the Apache License, Version 2.0 (the "License");
# you may not use this file except in compliance with the License.
# You may obtain a copy of the License at
#
#     http://www.apache.org/licenses/LICENSE-2.0
#
# Unless required by applicable law or agreed to in writing, software
# distributed under the License is distributed on an "AS IS" BASIS,
# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
# See the License for the specific language governing permissions and
# limitations under the License.

import torch
import torch.nn as nn

from detrex.layers.box_ops import box_cxcywh_to_xyxy
from detrex.modeling.losses import FocalLoss, GIoULoss, L1Loss
from detrex.modeling.matcher import FocalLossCost, GIoUCost, L1Cost, ModifedMatcher
from detrex.utils import get_world_size, is_dist_avail_and_initialized


class BaseCriterion(nn.Module):
    """Base criterion for calculating losses for DETR-like models.

    The process happens in two steps:
        1) we compute hungarian assignment between ground truth boxes and the outputs of the model
        2) we supervise each pair of matched ground-truth / prediction (supervise class and box)
    """

    def __init__(
        self,
        num_classes: int,
        matcher=ModifedMatcher(
            cost_class=FocalLossCost(
                alpha=0.25,
                gamma=2.0,
                weight=2.0,
            ),
            cost_bbox=L1Cost(weight=5.0),
            cost_giou=GIoUCost(weight=2.0),
        ),
        loss_class: nn.Module = FocalLoss(
            alpha=0.25,
            gamma=2.0,
            loss_weight=1.0,
        ),
        loss_bbox: nn.Module = L1Loss(loss_weight=5.0),
        loss_giou: nn.Module = GIoULoss(eps=1e-6, loss_weight=2.0),
    ):
        super().__init__()
        self.num_classes = num_classes
        self.matcher = matcher
        self.loss_class = loss_class
        self.loss_bbox = loss_bbox
        self.loss_giou = loss_giou

    def _get_src_permutation_idx(self, indices):
        # permute predictions following indices
        batch_idx = torch.cat([torch.full_like(src, i) for i, (src, _) in enumerate(indices)])
        src_idx = torch.cat([src for (src, _) in indices])
        return batch_idx, src_idx

    def _get_tgt_permutation_idx(self, indices):
        # permute targets following indices
        batch_idx = torch.cat([torch.full_like(tgt, i) for i, (_, tgt) in enumerate(indices)])
        tgt_idx = torch.cat([tgt for (_, tgt) in indices])
        return batch_idx, tgt_idx

    def calculate_class_loss(self, pred_logits, targets, indices, num_boxes):
        """
        Args:
            preds (torch.Tensor): The predicted logits with shape ``(bs, num_queries, num_classes)``.
            targets (dict):
            indices (list):
            num_boxes (int):
        """
        idx = self._get_src_permutation_idx(indices)
        target_classes_o = torch.cat([t["labels"][J] for t, (_, J) in zip(targets, indices)])
        target_classes = torch.full(
            pred_logits.shape[:2],
            self.num_classes,
            dtype=torch.int64,
            device=pred_logits.device,
        )
        target_classes[idx] = target_classes_o

        # Compute classification loss
        pred_logits = pred_logits.view(-1, self.num_classes)
        target_classes = target_classes.flatten()
        losses = self.loss_class(pred_logits, target_classes, avg_factor=num_boxes)
        return losses

    def calculate_bbox_loss(self, pred_boxes, targets, indices, num_boxes):
        idx = self._get_src_permutation_idx(indices)
        pred_boxes = pred_boxes[idx]
        target_boxes = torch.cat([t["boxes"][i] for t, (_, i) in zip(targets, indices)], dim=0)

        # Compute regression loss
        losses = self.loss_bbox(pred_boxes, target_boxes, avg_factor=num_boxes)
        return losses

    def calculate_giou_loss(self, pred_boxes, targets, indices, num_boxes):
        idx = self._get_src_permutation_idx(indices)
        pred_boxes = pred_boxes[idx]
        target_boxes = torch.cat([t["boxes"][i] for t, (_, i) in zip(targets, indices)], dim=0)

        # Convert box format to (x1, y1, x2, y2)
        pred_boxes = box_cxcywh_to_xyxy(pred_boxes)
        target_boxes = box_cxcywh_to_xyxy(target_boxes)

        # Compute iou loss
        losses = self.loss_giou(pred_boxes, target_boxes, avg_factor=num_boxes)
        return losses

    def forward(self, outputs, targets):
        output_without_aux = {k: v for k, v in outputs.items() if k != "aux_outputs"}

        # Collect preds and targets excluding aux_outputs for matcher
        pred_logits = output_without_aux["pred_logits"]
        pred_boxes = output_without_aux["pred_boxes"]
        target_labels_list = [v["labels"] for v in targets]
        target_boxes_list = [v["boxes"] for v in targets]

        # Retrieve the matching between the outputs of the last layer and the targets
        indices = self.matcher(
            pred_logits,
            pred_boxes,
            target_labels_list,
            target_boxes_list,
        )

        # Compute the average number of target boxes accross all nodes, for normalization purposes
        num_boxes = sum(len(t["labels"]) for t in targets)
        num_boxes = torch.as_tensor(
            [num_boxes], dtype=torch.float, device=next(iter(outputs.values())).device
        )
        if is_dist_avail_and_initialized():
            torch.distributed.all_reduce(num_boxes)
        num_boxes = torch.clamp(num_boxes / get_world_size(), min=1).item()

        # Compute all losses for DETR-like models
        losses = {}
        losses["loss_class"] = self.calculate_class_loss(pred_logits, targets, indices, num_boxes)
        losses["loss_bbox"] = self.calculate_bbox_loss(pred_boxes, targets, indices, num_boxes)
        losses["loss_giou"] = self.calculate_giou_loss(pred_boxes, targets, indices, num_boxes)

        # In case of auxiliary losses, we repeat this process with the output of each intermediate layer.
        if "aux_outputs" in outputs:
            for i, aux_output in enumerate(outputs["aux_outputs"]):
                aux_pred_logits = aux_output["pred_logits"]
                aux_pred_boxes = aux_output["pred_boxes"]
                indices = self.matcher(
                    aux_pred_logits, aux_pred_boxes, target_labels_list, target_boxes_list
                )
                losses["loss_class" + f"_{i}"] = self.calculate_class_loss(
                    aux_pred_logits, targets, indices, num_boxes
                )
                losses["loss_bbox" + f"_{i}"] = self.calculate_bbox_loss(
                    aux_pred_boxes, targets, indices, num_boxes
                )
                losses["loss_giou" + f"_{i}"] = self.calculate_giou_loss(
                    aux_pred_boxes, targets, indices, num_boxes
                )

        return losses