File size: 8,679 Bytes
c705408
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
# %BANNER_BEGIN%
# ---------------------------------------------------------------------
# %COPYRIGHT_BEGIN%
#
#  Magic Leap, Inc. ("COMPANY") CONFIDENTIAL
#
#  Unpublished Copyright (c) 2020
#  Magic Leap, Inc., All Rights Reserved.
#
# NOTICE:  All information contained herein is, and remains the property
# of COMPANY. The intellectual and technical concepts contained herein
# are proprietary to COMPANY and may be covered by U.S. and Foreign
# Patents, patents in process, and are protected by trade secret or
# copyright law.  Dissemination of this information or reproduction of
# this material is strictly forbidden unless prior written permission is
# obtained from COMPANY.  Access to the source code contained herein is
# hereby forbidden to anyone except current COMPANY employees, managers
# or contractors who have executed Confidentiality and Non-disclosure
# agreements explicitly covering such access.
#
# The copyright notice above does not evidence any actual or intended
# publication or disclosure  of  this source code, which includes
# information that is confidential and/or proprietary, and is a trade
# secret, of  COMPANY.   ANY REPRODUCTION, MODIFICATION, DISTRIBUTION,
# PUBLIC  PERFORMANCE, OR PUBLIC DISPLAY OF OR THROUGH USE  OF THIS
# SOURCE CODE  WITHOUT THE EXPRESS WRITTEN CONSENT OF COMPANY IS
# STRICTLY PROHIBITED, AND IN VIOLATION OF APPLICABLE LAWS AND
# INTERNATIONAL TREATIES.  THE RECEIPT OR POSSESSION OF  THIS SOURCE
# CODE AND/OR RELATED INFORMATION DOES NOT CONVEY OR IMPLY ANY RIGHTS
# TO REPRODUCE, DISCLOSE OR DISTRIBUTE ITS CONTENTS, OR TO MANUFACTURE,
# USE, OR SELL ANYTHING THAT IT  MAY DESCRIBE, IN WHOLE OR IN PART.
#
# %COPYRIGHT_END%
# ----------------------------------------------------------------------
# %AUTHORS_BEGIN%
#
#  Originating Authors: Paul-Edouard Sarlin
#
# %AUTHORS_END%
# --------------------------------------------------------------------*/
# %BANNER_END%

# Adapted by Remi Pautrat, Philipp Lindenberger

import torch
from kornia.color import rgb_to_grayscale
from torch import nn

from .utils import Extractor


def simple_nms(scores, nms_radius: int):
    """Fast Non-maximum suppression to remove nearby points"""
    assert nms_radius >= 0

    def max_pool(x):
        return torch.nn.functional.max_pool2d(
            x, kernel_size=nms_radius * 2 + 1, stride=1, padding=nms_radius
        )

    zeros = torch.zeros_like(scores)
    max_mask = scores == max_pool(scores)
    for _ in range(2):
        supp_mask = max_pool(max_mask.float()) > 0
        supp_scores = torch.where(supp_mask, zeros, scores)
        new_max_mask = supp_scores == max_pool(supp_scores)
        max_mask = max_mask | (new_max_mask & (~supp_mask))
    return torch.where(max_mask, scores, zeros)


def top_k_keypoints(keypoints, scores, k):
    if k >= len(keypoints):
        return keypoints, scores
    scores, indices = torch.topk(scores, k, dim=0, sorted=True)
    return keypoints[indices], scores


def sample_descriptors(keypoints, descriptors, s: int = 8):
    """Interpolate descriptors at keypoint locations"""
    b, c, h, w = descriptors.shape
    keypoints = keypoints - s / 2 + 0.5
    keypoints /= torch.tensor(
        [(w * s - s / 2 - 0.5), (h * s - s / 2 - 0.5)],
    ).to(
        keypoints
    )[None]
    keypoints = keypoints * 2 - 1  # normalize to (-1, 1)
    args = {"align_corners": True} if torch.__version__ >= "1.3" else {}
    descriptors = torch.nn.functional.grid_sample(
        descriptors, keypoints.view(b, 1, -1, 2), mode="bilinear", **args
    )
    descriptors = torch.nn.functional.normalize(
        descriptors.reshape(b, c, -1), p=2, dim=1
    )
    return descriptors


class SuperPoint(Extractor):
    """SuperPoint Convolutional Detector and Descriptor

    SuperPoint: Self-Supervised Interest Point Detection and
    Description. Daniel DeTone, Tomasz Malisiewicz, and Andrew
    Rabinovich. In CVPRW, 2019. https://arxiv.org/abs/1712.07629

    """

    default_conf = {
        "descriptor_dim": 256,
        "nms_radius": 4,
        "max_num_keypoints": None,
        "detection_threshold": 0.0005,
        "remove_borders": 4,
    }

    preprocess_conf = {
        "resize": 1024,
    }

    required_data_keys = ["image"]

    def __init__(self, **conf):
        super().__init__(**conf)  # Update with default configuration.
        self.relu = nn.ReLU(inplace=True)
        self.pool = nn.MaxPool2d(kernel_size=2, stride=2)
        c1, c2, c3, c4, c5 = 64, 64, 128, 128, 256

        self.conv1a = nn.Conv2d(1, c1, kernel_size=3, stride=1, padding=1)
        self.conv1b = nn.Conv2d(c1, c1, kernel_size=3, stride=1, padding=1)
        self.conv2a = nn.Conv2d(c1, c2, kernel_size=3, stride=1, padding=1)
        self.conv2b = nn.Conv2d(c2, c2, kernel_size=3, stride=1, padding=1)
        self.conv3a = nn.Conv2d(c2, c3, kernel_size=3, stride=1, padding=1)
        self.conv3b = nn.Conv2d(c3, c3, kernel_size=3, stride=1, padding=1)
        self.conv4a = nn.Conv2d(c3, c4, kernel_size=3, stride=1, padding=1)
        self.conv4b = nn.Conv2d(c4, c4, kernel_size=3, stride=1, padding=1)

        self.convPa = nn.Conv2d(c4, c5, kernel_size=3, stride=1, padding=1)
        self.convPb = nn.Conv2d(c5, 65, kernel_size=1, stride=1, padding=0)

        self.convDa = nn.Conv2d(c4, c5, kernel_size=3, stride=1, padding=1)
        self.convDb = nn.Conv2d(
            c5, self.conf.descriptor_dim, kernel_size=1, stride=1, padding=0
        )

        url = "https://github.com/cvg/LightGlue/releases/download/v0.1_arxiv/superpoint_v1.pth"  # noqa
        self.load_state_dict(torch.hub.load_state_dict_from_url(url,model_dir='./LightGlue/ckpts/',file_name='superpoint_v1.pth'))

        if self.conf.max_num_keypoints is not None and self.conf.max_num_keypoints <= 0:
            raise ValueError("max_num_keypoints must be positive or None")

    def forward(self, data: dict) -> dict:
        """Compute keypoints, scores, descriptors for image"""
        for key in self.required_data_keys:
            assert key in data, f"Missing key {key} in data"
        image = data["image"]
        if image.shape[1] == 3:
            image = rgb_to_grayscale(image)

        # Shared Encoder
        x = self.relu(self.conv1a(image))
        x = self.relu(self.conv1b(x))
        x = self.pool(x)
        x = self.relu(self.conv2a(x))
        x = self.relu(self.conv2b(x))
        x = self.pool(x)
        x = self.relu(self.conv3a(x))
        x = self.relu(self.conv3b(x))
        x = self.pool(x)
        x = self.relu(self.conv4a(x))
        x = self.relu(self.conv4b(x))

        # Compute the dense keypoint scores
        cPa = self.relu(self.convPa(x))
        scores = self.convPb(cPa)
        scores = torch.nn.functional.softmax(scores, 1)[:, :-1]
        b, _, h, w = scores.shape
        scores = scores.permute(0, 2, 3, 1).reshape(b, h, w, 8, 8)
        scores = scores.permute(0, 1, 3, 2, 4).reshape(b, h * 8, w * 8)
        scores = simple_nms(scores, self.conf.nms_radius)

        # Discard keypoints near the image borders
        if self.conf.remove_borders:
            pad = self.conf.remove_borders
            scores[:, :pad] = -1
            scores[:, :, :pad] = -1
            scores[:, -pad:] = -1
            scores[:, :, -pad:] = -1

        # Extract keypoints
        best_kp = torch.where(scores > self.conf.detection_threshold)
        scores = scores[best_kp]

        # Separate into batches
        keypoints = [
            torch.stack(best_kp[1:3], dim=-1)[best_kp[0] == i] for i in range(b)
        ]
        scores = [scores[best_kp[0] == i] for i in range(b)]

        # Keep the k keypoints with highest score
        if self.conf.max_num_keypoints is not None:
            keypoints, scores = list(
                zip(
                    *[
                        top_k_keypoints(k, s, self.conf.max_num_keypoints)
                        for k, s in zip(keypoints, scores)
                    ]
                )
            )

        # Convert (h, w) to (x, y)
        keypoints = [torch.flip(k, [1]).float() for k in keypoints]

        # Compute the dense descriptors
        cDa = self.relu(self.convDa(x))
        descriptors = self.convDb(cDa)
        descriptors = torch.nn.functional.normalize(descriptors, p=2, dim=1)

        # Extract descriptors
        descriptors = [
            sample_descriptors(k[None], d[None], 8)[0]
            for k, d in zip(keypoints, descriptors)
        ]

        return {
            "keypoints": torch.stack(keypoints, 0),
            "keypoint_scores": torch.stack(scores, 0),
            "descriptors": torch.stack(descriptors, 0).transpose(-1, -2).contiguous(),
        }