videos_to_colmap.py 18.5 KB
Newer Older
Clement Pinard's avatar
Clement Pinard committed
1
2
3
4
5
6
from colmap_util import read_model as rm, database as db
import anafi_metadata as am
from wrappers import FFMpeg, PDraw
from sklearn.cluster import KMeans
from sklearn.metrics import pairwise_distances_argmin_min
from argparse import ArgumentParser, ArgumentDefaultsHelpFormatter
nicolas's avatar
nicolas committed
7
from edit_exif import set_gps_location
Clement Pinard's avatar
Clement Pinard committed
8
9
10
11
12
from path import Path
import pandas as pd
import numpy as np
from pyproj import Proj
from tqdm import tqdm
13
import tempfile
Clement Pinard's avatar
Clement Pinard committed
14
15
16
17
18
19
20
21
22

parser = ArgumentParser(description='Take all the drone videos of a folder and put the frame '
                                    'location in a COLMAP file for vizualisation',
                        formatter_class=ArgumentDefaultsHelpFormatter)

parser.add_argument('--video_folder', metavar='DIR',
                    help='path to videos', type=Path)
parser.add_argument('--system', default='epsg:2154')
parser.add_argument('--centroid_path', default=None)
Clément Pinard's avatar
Clément Pinard committed
23
parser.add_argument('--colmap_img_root', metavar='DIR', type=Path)
Clement Pinard's avatar
Clement Pinard committed
24
25
26
27
28
29
30
parser.add_argument('--output_format', metavar='EXT', default="bin")
parser.add_argument('--vid_ext', nargs='+', default=[".mp4", ".MP4"])
parser.add_argument('--pic_ext', nargs='+', default=[".jpg", ".JPG", ".png", ".PNG"])
parser.add_argument('--nw', default='',
                    help="native-wrapper.sh file location")
parser.add_argument('--fps', default=1, type=int,
                    help="framerate at which videos will be scanned WITH reconstruction")
31
parser.add_argument('--total_frames', default=200, type=int)
Clement Pinard's avatar
Clement Pinard committed
32
33
34
parser.add_argument('--orientation_weight', default=1, type=float)
parser.add_argument('--resolution_weight', default=1, type=float)
parser.add_argument('--save_space', action="store_true")
Clément Pinard's avatar
Clément Pinard committed
35
36
parser.add_argument('--thorough_db', type=Path)
parser.add_argument('-v', '--verbose', action="count", default=0)
Clement Pinard's avatar
Clement Pinard committed
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55


def world_coord_from_frame(frame_qvec, frame_tvec):
    '''
    frame_qvec is written in the NED system (north east down)
    frame_tvec is already is the world system (east norht up)
    '''
    world2NED = np.float32([[0, 1, 0],
                            [1, 0, 0],
                            [0, 0, -1]])
    NED2cam = np.float32([[0, 1, 0],
                          [0, 0, 1],
                          [1, 0, 0]])
    world2cam = NED2cam @ rm.qvec2rotmat(frame_qvec).T @ world2NED
    cam_tvec = - world2cam  @ frame_tvec
    cam_qvec = rm.rotmat2qvec(world2cam)
    return cam_qvec, cam_tvec


nicolas's avatar
nicolas committed
56
57
58
59
60
61
def set_gps(frames_list, metadata, image_path):
    for frame in frames_list:
        relative = str(frame.relpath(image_path))
        row = metadata[metadata["image_path"] == relative]
        if len(row) > 0:
            row = row.iloc[0]
62
63
64
65
66
            if row["location_valid"]:
                set_gps_location(frame,
                                 lat=row["location_latitude"],
                                 lng=row["location_longitude"],
                                 altitude=row["location_altitude"])
nicolas's avatar
nicolas committed
67
68
69
70
71
72
73
74
75
76
77
78
79


def get_georef(metadata):
    relevant_data = metadata[["location_valid", "image_path", "x", "y", "z"]]
    path_list = []
    georef_list = []
    for _, (gps, path, x, y, alt) in relevant_data.iterrows():
        path_list.append(path)
        if gps == 1:
            georef_list.append("{} {} {} {}\n".format(path, x, y, alt))
    return georef_list, path_list


Clement Pinard's avatar
Clement Pinard committed
80
def optimal_sample(metadata, num_frames, orientation_weight, resolution_weight):
81
    valid_metadata = metadata[~metadata["sampled"]].dropna()
Clément Pinard's avatar
Clément Pinard committed
82
83
    if len(valid_metadata) == 0:
        return metadata
84
85
86
87
88
89
    XYZ = valid_metadata[["x", "y", "z"]].values
    axis_angle = valid_metadata[["frame_quat_x", "frame_quat_y", "frame_quat_z"]].values
    if True in valid_metadata["indoor"].unique():
        # We have indoor videos, without absolute positions. We assume each video is very far
        # from the other ones. As such we will have an optimal subsampling of each video
        # It won't leverage video proximity from each other but it's better than nothing
Clement Pinard's avatar
Clement Pinard committed
90
        diameter = (XYZ.max(axis=0) - XYZ.min(axis=0))
91
92
93
94
        indoor_videos = valid_metadata.loc[valid_metadata["indoor"]]["video"].unique()
        new_centroids = 2 * diameter * np.linspace(0, 10, len(indoor_videos)).reshape(-1, 1)
        for centroid, v in zip(new_centroids, indoor_videos):
            video_index = (valid_metadata["video"] == v).values
Clement Pinard's avatar
Clement Pinard committed
95
96
            XYZ[video_index] += centroid

97
    frame_size = valid_metadata["video_quality"].values
Clement Pinard's avatar
Clement Pinard committed
98
99
100
101
102
103
104
105
    weighted_point_cloud = np.concatenate([XYZ, orientation_weight * axis_angle], axis=1)

    if resolution_weight == 0:
        weights = None
    else:
        weights = frame_size ** resolution_weight
    km = KMeans(n_clusters=num_frames).fit(weighted_point_cloud, sample_weight=weights)
    closest, _ = pairwise_distances_argmin_min(km.cluster_centers_, weighted_point_cloud)
106
    metadata.at[valid_metadata.index[closest], "sampled"] = True
Clement Pinard's avatar
Clement Pinard committed
107
108
109
    return metadata


110
def register_new_cameras(cameras_dataframe, database, camera_dict):
Clement Pinard's avatar
Clement Pinard committed
111
    camera_ids = []
Clément Pinard's avatar
Clément Pinard committed
112
113
114
115
116
    for _, row in cameras_dataframe.iterrows():
        w, h, hfov, vfov, camera_model = row.reindex(["width", "height", "picture_hfov", "picture_vfov", "camera_model"])
        prior_focal_length = False
        single_focal = ('SIMPLE' in camera_model) or ('RADIAL' in camera_model)
        if hfov != 0:
117
            fx = w / (2 * np.tan(hfov * np.pi/360))
Clément Pinard's avatar
Clément Pinard committed
118
119
120
121
122
            # If the model is not single focal, only knowing hfov is not enough, you also need to know vfov
            prior_focal_length = single_focal
        else:
            fx = w / 2  # As if hfov was 90 degrees
        if vfov != 0:
123
            fy = h / (2 * np.tan(vfov * np.pi/360))
Clément Pinard's avatar
Clément Pinard committed
124
            prior_focal_length = True
125
        else:
Clément Pinard's avatar
Clément Pinard committed
126
            fy = w / 2  # As if vfov was 90 degrees
127
128
        model_id = rm.CAMERA_MODEL_NAMES[camera_model].model_id
        num_params = rm.CAMERA_MODEL_NAMES[camera_model].num_params
Clément Pinard's avatar
Clément Pinard committed
129
130
131
132
133
        if ('SIMPLE' in camera_model) or ('RADIAL' in camera_model):
            params = np.array([fx, w/2, h/2] + [0] * (num_params - 3))
        else:
            params = np.array([fx, fy, w/2, h/2] + [0] * (num_params - 4))
        db_id = database.add_camera(model_id, int(w), int(h), params, prior_focal_length=prior_focal_length)
Clement Pinard's avatar
Clement Pinard committed
134
135
        camera_ids.append(db_id)
        camera_dict[db_id] = rm.Camera(id=db_id,
136
                                       model=camera_model,
Clement Pinard's avatar
Clement Pinard committed
137
138
139
140
141
142
143
144
                                       width=int(w),
                                       height=int(h),
                                       params=params)
    ids_series = pd.Series(camera_ids)
    return cameras_dataframe.set_index(ids_series)


def process_video_folder(videos_list, existing_pictures, output_video_folder, image_path, system, centroid,
Clément Pinard's avatar
Clément Pinard committed
145
                         thorough_db, fps=1, total_frames=500, orientation_weight=1, resolution_weight=1,
146
147
                         output_colmap_format="bin", save_space=False, include_lowfps_thorough=False,
                         max_sequence_length=1000, **env):
Clement Pinard's avatar
Clement Pinard committed
148
149
150
151
152
    proj = Proj(system)
    final_metadata = []
    video_output_folders = {}
    images = {}
    colmap_cameras = {}
153
    tempfile_database = Path(tempfile.NamedTemporaryFile().name)
Clément Pinard's avatar
Clément Pinard committed
154
155
    if thorough_db.isfile():
        thorough_db.copy(thorough_db.stripext() + "_backup.db")
Clement Pinard's avatar
Clement Pinard committed
156
    path_lists_output = {}
157
    database = db.COLMAPDatabase.connect(thorough_db)
Clement Pinard's avatar
Clement Pinard committed
158
159
160
    database.create_tables()

    print("extracting metadata for {} videos...".format(len(videos_list)))
161
    videos_summary = {"anafi": {"indoor": 0, "outdoor": 0}, "generic": 0}
Clement Pinard's avatar
Clement Pinard committed
162
    for v in tqdm(videos_list):
163
        width, height, framerate, num_frames = env["ffmpeg"].get_size_and_framerate(v)
Clement Pinard's avatar
Clement Pinard committed
164
165
166
167
        video_output_folder = output_video_folder / "{}x{}".format(width, height) / v.namebase
        video_output_folder.makedirs_p()
        video_output_folders[v] = video_output_folder

168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
        try:
            metadata = am.extract_metadata(v.parent, v, env["pdraw"], proj,
                                           width, height, framerate)
            metadata["model"] = "anafi"
            metadata["camera_model"] = "PINHOLE"
            if metadata["indoor"].iloc[0]:
                videos_summary["anafi"]["indoor"] += 1
            else:
                videos_summary["anafi"]["outdoor"] += 1
                raw_positions = metadata[["x", "y", "z"]]
                if centroid is None:
                    '''No centroid (possibly because there was no georeferenced lidar model in the first place)
                    set it as the first valid GPS position of the first outdoor video'''
                    centroid = raw_positions[metadata["location_valid"] == 1].iloc[0].values
                zero_centered_positions = raw_positions.values - centroid
Clément Pinard's avatar
bug fix    
Clément Pinard committed
183
184
185
186
                radius = np.max(np.abs(zero_centered_positions))
                if radius > 1000:
                    print("Warning, your positions coordinates are most likely too high, have you configured the right GPS system ?")
                    print("It should be the same as the one used for the Lidar point cloud")
187
188
189
190
191
192
193
194
195
196
197
198
199
200
                metadata["x"], metadata["y"], metadata["z"] = zero_centered_positions.transpose()
        except Exception:
            # No metadata found, construct a simpler dataframe without location
            metadata = pd.DataFrame({"video": [v] * num_frames})
            metadata["height"] = height
            metadata["width"] = width
            metadata["framerate"] = framerate
            metadata["video_quality"] = height * width / framerate
            metadata['frame'] = metadata.index + 1
            # timestemp is in microseconds
            metadata['time'] = 1e6 * metadata.index / framerate
            metadata['indoor'] = True
            metadata['location_valid'] = 0
            metadata["model"] = "generic"
Clément Pinard's avatar
Clément Pinard committed
201
202
203
            metadata["camera_model"] = "PINHOLE"
            metadata["picture_hfov"] = 0
            metadata["picture_vfov"] = 0
Clément Pinard's avatar
bug fix    
Clément Pinard committed
204
205
206
207
            metadata["frame_quat_w"] = np.NaN
            metadata["frame_quat_x"] = np.NaN
            metadata["frame_quat_y"] = np.NaN
            metadata["frame_quat_z"] = np.NaN
Clément Pinard's avatar
Clément Pinard committed
208
209
210
            metadata["x"] = np.NaN
            metadata["y"] = np.NaN
            metadata["z"] = np.NaN
211
212
213
214
215
            videos_summary["generic"] += 1
        if include_lowfps_thorough:
            by_time = metadata.set_index(pd.to_datetime(metadata["time"], unit="us"))
            by_time_lowfps = by_time.resample("{:.3f}S".format(1/fps)).first()
            metadata["sampled"] = by_time["time"].isin(by_time_lowfps["time"]).values
Clément Pinard's avatar
Clément Pinard committed
216
        else:
217
218
            metadata["sampled"] = False
        final_metadata.append(metadata)
Clement Pinard's avatar
Clement Pinard committed
219
    final_metadata = pd.concat(final_metadata, ignore_index=True)
220
221
222
    print("{} outdoor anafi videos".format(videos_summary["anafi"]["outdoor"]))
    print("{} indoor anafi videos".format(videos_summary["anafi"]["indoor"]))
    print("{} generic videos".format(videos_summary["generic"]))
Clement Pinard's avatar
Clement Pinard committed
223
224
225

    print("{} frames in total".format(len(final_metadata)))

226
227
228
    cam_fields = ["width", "height", "framerate", "picture_hfov", "picture_vfov", "camera_model"]
    cameras_dataframe = final_metadata[final_metadata["model"] == "anafi"][cam_fields].drop_duplicates()
    cameras_dataframe = register_new_cameras(cameras_dataframe, database, colmap_cameras)
Clement Pinard's avatar
Clement Pinard committed
229
230
231
    final_metadata["camera_id"] = 0
    for cam_id, row in cameras_dataframe.iterrows():
        final_metadata.loc[(final_metadata[cam_fields] == row).all(axis=1), "camera_id"] = cam_id
232
233
234
    if any(final_metadata["model"] == "generic"):
        print("Undefined remaining cameras, assigning generic models to them")
        generic_frames = final_metadata[final_metadata["model"] == "generic"]
Clément Pinard's avatar
Clément Pinard committed
235
236
        generic_cam_fields = cam_fields + ["video"]
        generic_cameras_dataframe = generic_frames[generic_cam_fields]
237
238
239
240
241
242
        fixed_camera = True
        if fixed_camera:
            generic_cameras_dataframe = generic_cameras_dataframe.drop_duplicates()
        generic_cameras_dataframe = register_new_cameras(generic_cameras_dataframe, database, colmap_cameras)
        if fixed_camera:
            for cam_id, row in generic_cameras_dataframe.iterrows():
Clément Pinard's avatar
Clément Pinard committed
243
                final_metadata.loc[(final_metadata[generic_cam_fields] == row).all(axis=1), "camera_id"] = cam_id
244
245
246
247
248
249
250
        else:
            final_metadata.loc[generic_frames.index, "camera_id"] = generic_cameras_dataframe.index
        cameras_dataframe = cameras_dataframe.append(generic_cameras_dataframe)
    print("Cameras : ")
    print(cameras_dataframe)

    to_extract = total_frames - len(existing_pictures) - sum(final_metadata["sampled"])
Clement Pinard's avatar
Clement Pinard committed
251

Clément Pinard's avatar
Clément Pinard committed
252
    if to_extract <= 0:
253
        pass
Clément Pinard's avatar
Clément Pinard committed
254
    elif to_extract < len(final_metadata):
Clement Pinard's avatar
Clement Pinard committed
255
256
257
258
259
260
        print("subsampling based on K-Means, to get {}"
              " frames from videos, for a total of {} frames".format(to_extract, total_frames))
        final_metadata = optimal_sample(final_metadata, total_frames - len(existing_pictures),
                                        orientation_weight,
                                        resolution_weight)
        print("Done.")
Clément Pinard's avatar
Clément Pinard committed
261
262
    else:
        final_metadata["sampled"] = True
Clement Pinard's avatar
Clement Pinard committed
263

264
    print("Constructing COLMAP model with {:,} frames".format(sum(final_metadata["sampled"])))
Clement Pinard's avatar
Clement Pinard committed
265

266
267
268
269
    database.commit()
    thorough_db.copy(tempfile_database)
    temp_database = db.COLMAPDatabase.connect(tempfile_database)

Clement Pinard's avatar
Clement Pinard committed
270
    final_metadata["image_path"] = ""
271
272
    final_metadata["db_id"] = -1
    for current_id, row in tqdm(final_metadata.iterrows(), total=len(final_metadata)):
Clement Pinard's avatar
Clement Pinard committed
273
274
275
276
277
        video = row["video"]
        frame = row["frame"]
        camera_id = row["camera_id"]
        current_image_path = video_output_folders[video].relpath(image_path) / video.namebase + "_{:05d}.jpg".format(frame)

278
279
280
        final_metadata.at[current_id, "image_path"] = current_image_path
        db_image_id = temp_database.add_image(current_image_path, int(camera_id))
        final_metadata.at[current_id, "db_id"] = db_image_id
Clement Pinard's avatar
Clement Pinard committed
281
282
283
284
285
286

        if row["sampled"]:
            frame_qvec = row[["frame_quat_w",
                              "frame_quat_x",
                              "frame_quat_y",
                              "frame_quat_z"]].values
287
288
            if True in pd.isnull(frame_qvec):
                frame_qvec = np.array([1, 0, 0, 0])
nicolas's avatar
nicolas committed
289
290
            x, y, z = row[["x", "y", "z"]]
            frame_tvec = np.array([x, y, z])
Clement Pinard's avatar
Clement Pinard committed
291
292
293
294
295
296
            if row["location_valid"]:
                frame_gps = row[["location_longitude", "location_latitude", "location_altitude"]]
            else:
                frame_gps = np.full(3, np.NaN)

            world_qvec, world_tvec = world_coord_from_frame(frame_qvec, frame_tvec)
297
298
299
300
            database.add_image(current_image_path, int(camera_id), prior_t=frame_gps, image_id=db_image_id)
            images[db_image_id] = rm.Image(id=db_image_id, qvec=world_qvec, tvec=world_tvec,
                                           camera_id=camera_id, name=current_image_path,
                                           xys=[], point3D_ids=[])
Clement Pinard's avatar
Clement Pinard committed
301
302
303

    database.commit()
    database.close()
304
305
    temp_database.commit()
    temp_database.close()
Clement Pinard's avatar
Clement Pinard committed
306
307
308
    rm.write_model(colmap_cameras, images, {}, output_video_folder, "." + output_colmap_format)
    print("COLMAP model created")

nicolas's avatar
nicolas committed
309
310
311
312
    thorough_georef, thorough_paths = get_georef(final_metadata[final_metadata["sampled"]])
    path_lists_output["thorough"] = {}
    path_lists_output["thorough"]["frames"] = thorough_paths
    path_lists_output["thorough"]["georef"] = thorough_georef
Clement Pinard's avatar
Clement Pinard committed
313
314
315
316
317

    print("Extracting frames from videos")

    for v in tqdm(videos_list):
        video_metadata = final_metadata[final_metadata["video"] == v]
nicolas's avatar
nicolas committed
318
        by_time = video_metadata.set_index(pd.to_datetime(video_metadata["time"], unit="us"))
Clement Pinard's avatar
Clement Pinard committed
319
320
        video_folder = video_output_folders[v]
        video_metadata.to_csv(video_folder/"metadata.csv")
nicolas's avatar
nicolas committed
321
322
323
324
325
        path_lists_output[v] = {}
        video_metadata_1fps = by_time.resample("{:.3f}S".format(1/fps)).first()
        georef, frame_paths = get_georef(video_metadata_1fps)
        path_lists_output[v]["frames_lowfps"] = frame_paths
        path_lists_output[v]["georef_lowfps"] = georef
Clément Pinard's avatar
Clément Pinard committed
326
        num_chunks = len(video_metadata) // max_sequence_length + 1
327
328
329
330
331
332
333
        chunks = [list(frames) for frames in np.array_split(video_metadata["image_path"],
                                                            num_chunks)]
        # Add some overlap between chunks, in order to ease the model merging afterwards
        for chunk, next_chunk in zip(chunks, chunks[1:]):
            chunk.extend(next_chunk[:10])
        path_lists_output[v]["frames_full"] = chunks

Clement Pinard's avatar
Clement Pinard committed
334
        if save_space:
Clément Pinard's avatar
Clément Pinard committed
335
336
            frame_ids = set(video_metadata[video_metadata["sampled"]]["frame"].values) | \
                set(video_metadata_1fps["frame"].values)
337
            frame_ids = sorted(list(frame_ids))
Clément Pinard's avatar
Clément Pinard committed
338
            if len(frame_ids) > 0:
nicolas's avatar
nicolas committed
339
                extracted_frames = env["ffmpeg"].extract_specific_frames(v, video_folder, frame_ids)
Clement Pinard's avatar
Clement Pinard committed
340
        else:
nicolas's avatar
nicolas committed
341
342
            extracted_frames = env["ffmpeg"].extract_images(v, video_folder)
        set_gps(extracted_frames, video_metadata, image_path)
Clement Pinard's avatar
Clement Pinard committed
343
344
345
346
347
348
349
350

    return path_lists_output, video_output_folders


if __name__ == '__main__':
    args = parser.parse_args()
    env = vars(args)
    env["videos_list"] = sum((list(args.video_folder.walkfiles('*{}'.format(ext))) for ext in args.vid_ext), [])
Clément Pinard's avatar
Clément Pinard committed
351
    output_video_folder = args.colmap_img_root / "Videos"
Clement Pinard's avatar
Clement Pinard committed
352
    output_video_folder.makedirs_p()
Clément Pinard's avatar
Clément Pinard committed
353
    env["image_path"] = args.colmap_img_root
Clement Pinard's avatar
Clement Pinard committed
354
    env["output_video_folder"] = output_video_folder
Clément Pinard's avatar
Clément Pinard committed
355
    env["existing_pictures"] = sum((list(args.colmap_img_root.walkfiles('*{}'.format(ext))) for ext in args.pic_ext), [])
Clément Pinard's avatar
Clément Pinard committed
356
357
358
    env["pdraw"] = PDraw(args.nw, verbose=args.verbose)
    env["ffmpeg"] = FFMpeg(verbose=args.verbose)
    env["output_colmap_format"] = args.output_format
Clement Pinard's avatar
Clement Pinard committed
359
360
361
362
363
364
365
366
367

    if args.centroid_path is not None:
        centroid = np.loadtxt(args.centroid_path)
    else:
        centroid = np.zeros(3)
    env["centroid"] = centroid
    lists, extracted_video_folders = process_video_folder(**env)

    if lists is not None:
Clément Pinard's avatar
Clément Pinard committed
368
369
370
        with open(args.colmap_img_root/"video_frames_for_thorough_scan.txt", "w") as f:
            f.write("\n".join(lists["thorough"]["frames"]) + "\n")
        with open(args.colmap_img_root/"georef.txt", "w") as f:
Clément Pinard's avatar
Clément Pinard committed
371
            f.write("\n".join(lists["thorough"]["georef"]))
Clement Pinard's avatar
Clement Pinard committed
372
        for v in env["videos_list"]:
Clément Pinard's avatar
Clément Pinard committed
373
374
375
376
377
378
379
380
381
            video_folder = extracted_video_folders[v]
            with open(video_folder / "lowfps.txt", "w") as f:
                f.write("\n".join(lists[v]["frames_lowfps"]) + "\n")
            with open(video_folder / "georef.txt", "w") as f:
                f.write("\n".join(lists["thorough"]["georef"]) + "\n")
                f.write("\n".join(lists[v]["georef_lowfps"]) + "\n")
            for j, l in enumerate(lists[v]["frames_full"]):
                with open(video_folder / "full_chunk_{}.txt".format(j), "w") as f:
                    f.write("\n".join(l) + "\n")