File size: 11,061 Bytes
0a82b18
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
import torch
import os
import torchvision.transforms as tfm
import py3_wget
from matching import BaseMatcher, THIRD_PARTY_DIR, WEIGHTS_DIR
from matching.utils import resize_to_divisible, add_to_path

# add_to_path(THIRD_PARTY_DIR.joinpath("DeDoDe"))
# from DeDoDe import (
#     dedode_detector_L,
#     dedode_descriptor_B,
# )

add_to_path(THIRD_PARTY_DIR.joinpath("affine-steerers"))
from affine_steerers.utils import build_affine
from affine_steerers.matchers.dual_softmax_matcher import DualSoftMaxMatcher, MaxSimilarityMatcher
from affine_steerers import dedode_detector_L, dedode_descriptor_B, dedode_descriptor_G

class AffSteererMatcher(BaseMatcher):
    detector_path_L = WEIGHTS_DIR.joinpath("dedode_detector_C4_affsteerers.pth")

    descriptor_path_equi_G = WEIGHTS_DIR.joinpath("descriptor_aff_equi_G.pth")
    descriptor_path_steer_G = WEIGHTS_DIR.joinpath("descriptor_aff_steer_G.pth")

    descriptor_path_equi_B = WEIGHTS_DIR.joinpath("descriptor_aff_equi_B.pth")
    descriptor_path_steer_B = WEIGHTS_DIR.joinpath("descriptor_aff_steer_B.pth")

    steerer_path_equi_G = WEIGHTS_DIR.joinpath("steerer_aff_equi_G.pth")
    steerer_path_steer_G = WEIGHTS_DIR.joinpath("steerer_aff_steer_G.pth")

    steerer_path_equi_B = WEIGHTS_DIR.joinpath("steerer_aff_equi_B.pth")
    steerer_path_steer_B = WEIGHTS_DIR.joinpath("steerer_aff_steer_B.pth")

    dino_patch_size = 14

    STEERER_TYPES = ["equi_G", "steer_G", "equi_B", "steer_B"]

    def __init__(
        self,
        device="cpu",
        max_num_keypoints=10_000,
        steerer_type="equi_G",
        match_threshold=0.01,
        *args,
        **kwargs,
    ):
        super().__init__(device, **kwargs)

        if self.device != "cuda": # only cuda devices work due to autocast in cuda in upstream.
            raise ValueError("Only device 'cuda' supported for AffineSteerers.")

        WEIGHTS_DIR.mkdir(exist_ok=True)

        self.steerer_type = steerer_type
        if self.steerer_type not in self.STEERER_TYPES:
            raise ValueError(f'unsupported type for aff-steerer: {steerer_type}. Must choose from {self.STEERER_TYPES}.')

        # download detector / descriptor / steerer

        self.download_weights()

        self.max_keypoints = max_num_keypoints
        self.threshold = match_threshold

        self.normalize = tfm.Normalize(mean=[0.485, 0.456, 0.406], std=[0.229, 0.224, 0.225])


        self.detector, self.descriptor, self.steerer, self.matcher = self.build_matcher()

    def download_weights(self):
        if not AffSteererMatcher.detector_path_L.exists():
            print("Downloading dedode_detector_C4.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/dedode_detector_C4.pth",
                AffSteererMatcher.detector_path_L,
            )

        # download descriptors
        if self.steerer_type == "equi_G" and not AffSteererMatcher.descriptor_path_equi_G.exists():
            print("Downloading descriptor_aff_equi_G.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/descriptor_aff_equi_G.pth",
                AffSteererMatcher.descriptor_path_equi_G,
            )

        if self.steerer_type == "steer_G" and not AffSteererMatcher.descriptor_path_steer_G.exists():
            print("Downloading descriptor_aff_steer_G.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/descriptor_aff_steer_G.pth",
                AffSteererMatcher.descriptor_path_steer_G,
            )

        if self.steerer_type == "equi_B" and not AffSteererMatcher.descriptor_path_equi_B.exists():
            print("Downloading descriptor_aff_equi_B.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/descriptor_aff_equi_B.pth",
                AffSteererMatcher.descriptor_path_equi_B,
            )

        if self.steerer_type == "steer_B" and not AffSteererMatcher.descriptor_path_steer_B.exists():
            print("Downloading descriptor_aff_steer_B.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/descriptor_aff_steer_B.pth",
                AffSteererMatcher.descriptor_path_steer_B,
            )

        # download steerers
        if self.steerer_type == "equi_G" and not AffSteererMatcher.steerer_path_equi_G.exists():
            print("Downloading steerer_aff_equi_G.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/steerer_aff_equi_G.pth",
                AffSteererMatcher.steerer_path_equi_G,
            )
        if self.steerer_type == "steer_G" and not AffSteererMatcher.steerer_path_steer_G.exists():
            print("Downloading steerer_aff_steer_G.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/steerer_aff_steer_G.pth",
                AffSteererMatcher.steerer_path_steer_G,
            )

        if self.steerer_type == "equi_B" and not AffSteererMatcher.steerer_path_equi_B.exists():
            print("Downloading steerer_aff_equi_B.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/steerer_aff_equi_B.pth",
                AffSteererMatcher.steerer_path_equi_B,
            )
        if self.steerer_type == "steer_B" and not AffSteererMatcher.steerer_path_steer_B.exists():
            print("Downloading steerer_aff_steer_B.pth")
            py3_wget.download_file(
                "https://github.com/georg-bn/affine-steerers/releases/download/weights/steerer_aff_steer_B.pth",
                AffSteererMatcher.steerer_path_steer_B,
            )

    def build_matcher(self):
        detector = dedode_detector_L(weights=torch.load(self.detector_path_L, map_location=self.device))

        if "G" in self.steerer_type:
            descriptor_path = self.descriptor_path_equi_G if 'equi' in self.steerer_type else self.descriptor_path_steer_G
            descriptor = dedode_descriptor_G(
                weights=torch.load(descriptor_path, map_location=self.device)
            )
        else:
            descriptor_path = self.descriptor_path_equi_B if 'equi' in self.steerer_type else self.descriptor_path_steer_B
            descriptor = dedode_descriptor_B(
                weights=torch.load(self.descriptor_path, map_location=self.device)
            )

        if "G" in self.steerer_type:
            steerer_path = self.steerer_path_equi_G if 'equi' in self.steerer_type else self.steerer_path_steer_G
        else:
            steerer_path = self.steerer_path_equi_B if 'equi' in self.steerer_type else self.steerer_path_steer_B

        assert steerer_path.exists(), f"could not find steerer weights at {steerer_path}. Please check that they exist."
        steerer = self.load_steerer(
                    steerer_path
                ).to(self.device).eval()

        steerer.use_prototype_affines = True

        if 'steer' not in self.steerer_type:
            steerer.prototype_affines = torch.stack(
                [
                    build_affine(
                        angle_1=0.,
                        dilation_1=1.,
                        dilation_2=1.,
                        angle_2=r * 2 * torch.pi / 8
                    )
                    for r in range(8)
                ],  # + ... more affines
                dim=0,
            ).to(self.device)

        matcher = MaxSimilarityMatcher(
            steerer=steerer,
            normalize=False,
            inv_temp=5,
            threshold=self.threshold
        )
        if self.device == "cpu":
            detector = detector.to(torch.float32)
            descriptor = descriptor.to(torch.float32)
            steerer = steerer.to(torch.float32)
        return detector, descriptor, steerer, matcher

    @staticmethod
    def load_steerer(steerer_path, checkpoint=False, prototypes=True, feat_dim=256):
        from affine_steerers.steerers import SteererSpread
        if checkpoint:
            sd = torch.load(steerer_path, map_location="cpu")["steerer"]
        else:
            sd = torch.load(steerer_path, map_location="cpu")

        nbr_prototypes = 0
        if prototypes and "prototype_affines" in sd:
            nbr_prototypes = sd["prototype_affines"].shape[0]

        steerer = SteererSpread(
            feat_dim=feat_dim,
            max_order=4,
            normalize=True,
            normalize_only_higher=False,
            fix_order_1_scalings=False,
            max_determinant_scaling=None,
            block_diag_rot=False,
            block_diag_optimal_scalings=False,
            learnable_determinant_scaling=True,
            learnable_basis=True,
            learnable_reference_direction=False,
            use_prototype_affines=prototypes and "prototype_affines" in sd,
            prototype_affines_init=[
                torch.eye(2)
                for i in range(nbr_prototypes)
            ]
        )
        steerer.load_state_dict(sd)
        return steerer

    def preprocess(self, img):
        # ensure that the img has the proper w/h to be compatible with patch sizes
        _, h, w = img.shape
        orig_shape = h, w
        img = resize_to_divisible(img, self.dino_patch_size)

        img = self.normalize(img).unsqueeze(0).to(self.device)
        return img, orig_shape

    def _forward(self, img0, img1):
        img0, img0_orig_shape = self.preprocess(img0)
        img1, img1_orig_shape = self.preprocess(img1)

        batch_0 = {"image": img0}
        detections_0 = self.detector.detect(batch_0, num_keypoints=self.max_keypoints)
        keypoints_0, P_0 = detections_0["keypoints"], detections_0["confidence"]

        batch_1 = {"image": img1}
        detections_1 = self.detector.detect(batch_1, num_keypoints=self.max_keypoints)
        keypoints_1, P_1 = detections_1["keypoints"], detections_1["confidence"]

        description_0 = self.descriptor.describe_keypoints(batch_0, keypoints_0)["descriptions"]
        description_1 = self.descriptor.describe_keypoints(batch_1, keypoints_1)["descriptions"]

        matches_0, matches_1, _ = self.matcher.match(
            keypoints_0,
            description_0,
            keypoints_1,
            description_1,
        )

        H0, W0, H1, W1 = *img0.shape[-2:], *img1.shape[-2:]
        mkpts0, mkpts1 = self.matcher.to_pixel_coords(matches_0, matches_1, H0, W0, H1, W1)

        # dedode sometimes requires reshaping an image to fit vit patch size evenly, so we need to
        # rescale kpts to the original img
        mkpts0 = self.rescale_coords(mkpts0, *img0_orig_shape, H0, W0)
        mkpts1 = self.rescale_coords(mkpts1, *img1_orig_shape, H1, W1)

        return mkpts0, mkpts1, keypoints_0, keypoints_1, description_0, description_1