File size: 27,432 Bytes
1456309 f8ab00f 1456309 84d96b2 1456309 9cf9ecf 84d96b2 7f616ca 1456309 e886fcc 29263f0 e886fcc 1456309 841ad13 29263f0 7f616ca a81f4d5 29263f0 325158c a81f4d5 325158c 29263f0 a81f4d5 e886fcc 29263f0 325158c 29263f0 325158c 29263f0 a81f4d5 29263f0 7f616ca a81f4d5 7f616ca a81f4d5 7f616ca a81f4d5 29263f0 a81f4d5 29263f0 7f616ca a81f4d5 29263f0 7f616ca 325158c a81f4d5 29263f0 a81f4d5 29263f0 eef0638 29263f0 eef0638 29263f0 a81f4d5 7f616ca a81f4d5 29263f0 a81f4d5 29263f0 a81f4d5 29263f0 a81f4d5 29263f0 a81f4d5 29263f0 a81f4d5 29263f0 9492094 a81f4d5 9492094 29263f0 9492094 a81f4d5 9492094 a81f4d5 29263f0 7f616ca 29263f0 a81f4d5 29263f0 a81f4d5 29263f0 7f616ca 29263f0 a81f4d5 7f616ca a81f4d5 7f616ca a81f4d5 619b161 a81f4d5 29263f0 a81f4d5 29263f0 a81f4d5 29263f0 9492094 29263f0 e886fcc 29263f0 9de6077 29263f0 84d96b2 |
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 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 |
---
license: cc-by-4.0
pretty_name: Pixels Point Polygons
size_categories:
- 100K<n<1M
task_categories:
- image-segmentation
- object-detection
tags:
- Aerial
- Environement
- Multimodal
- Earth Observation
- Image
- Lidar
- ALS
- pointcloud
- Building
- Polygon
- Vectorization
language:
- en
# configs:
# - config_name: all
# data_files:
# - split: train
# path: "data/224/annotations/annotations_all_train.json"
# - split: val
# path: "data/224/annotations/annotations_all_val.json"
# - split: test
# path: "data/224/annotations/annotations_all_test.json"
# - config_name: CH
# data_files:
# - split: train
# path: "data/224/annotations/annotations_CH_train.json"
# - split: val
# path: "data/224/annotations/annotations_CH_val.json"
# - split: test
# path: "data/224/annotations/annotations_CH_test.json"
# - config_name: NY
# data_files:
# - split: train
# path: "data/224/annotations/annotations_NY_train.json"
# - split: val
# path: "data/224/annotations/annotations_NY_val.json"
# - split: test
# path: "data/224/annotations/annotations_NY_test.json"
# - config_name: NZ
# data_files:
# - split: train
# path: "data/224/annotations/annotations_NZ_train.json"
# - split: val
# path: "data/224/annotations/annotations_NZ_val.json"
# - split: test
# path: "data/224/annotations/annotations_NZ_test.json"
---
<div align="center">
<h1 align="center">The P<sup>3</sup> Dataset: Pixels, Points and Polygons <br> for Multimodal Building Vectorization</h1>
<h3><align="center">Raphael Sulzer<sup>1,2</sup> Liuyun Duan<sup>1</sup>
Nicolas Girard<sup>1</sup> Florent Lafarge<sup>2</sup></a></h3>
<align="center"><sup>1</sup>LuxCarta Technology <br> <sup>2</sup>Centre Inria d'UniversitΓ© CΓ΄te d'Azur
<img src="./teaser.jpg" width=100% height=100%>
<b>Figure 1</b>: A view of our dataset of Zurich, Switzerland
</div>
## Table of Contents
- [Abstract](#abstract)
- [Highlights](#highlights)
- [Dataset](#dataset)
- [Pretrained model weights](#pretrained-model-weights)
- [Code](#code)
- [Citation](#citation)
- [Acknowledgements](#acknowledgements)
## Abstract
<div align="justify">
We present the P<sup>3</sup> dataset, a large-scale multimodal benchmark for building vectorization, constructed from aerial LiDAR point clouds, high-resolution aerial imagery, and vectorized 2D building outlines, collected across three continents. The dataset contains over 10 billion LiDAR points with decimeter-level accuracy and RGB images at a ground sampling distance of 25 cm. While many existing datasets primarily focus on the image modality, P<sup>3</sup> offers a complementary perspective by also incorporating dense 3D information. We demonstrate that LiDAR point clouds serve as a robust modality for predicting building polygons, both in hybrid and end-to-end learning frameworks. Moreover, fusing aerial LiDAR and imagery further improves accuracy and geometric quality of predicted polygons. The P<sup>3</sup> dataset is publicly available, along with code and pretrained weights of three state-of-the-art models for building polygon prediction at https://github.com/raphaelsulzer/PixelsPointsPolygons.
</div>
## Highlights
- A global, multimodal dataset of aerial images, aerial LiDAR point clouds and building outline polygons, available at [huggingface.co/datasets/rsi/PixelsPointsPolygons](https://huggingface.co/datasets/rsi/PixelsPointsPolygons)
- A library for training and evaluating state-of-the-art deep learning methods on the dataset, available at [github.com/raphaelsulzer/PixelsPointsPolygons](https://github.com/raphaelsulzer/PixelsPointsPolygons)
- Pretrained model weights, available at [huggingface.co/rsi/PixelsPointsPolygons](https://huggingface.co/rsi/PixelsPointsPolygons)
- A paper with an extensive experimental validation, available at [arxiv.org/abs/2505.15379](https://arxiv.org/abs/2505.15379)
## Dataset
### Overview
<div align="left">
<img src="./worldmap.jpg" width=60% height=50%>
</div>
### Download
The recommended and fastest way to download the dataset is to run
```
pip install huggingface_hub
python scripts/download_dataset.py --dataset-root $DATA_ROOT
```
Optionally you can also download the dataset by running
```
git lfs install
git clone https://huggingface.co/datasets/rsi/PixelsPointsPolygons $DATA_ROOT
```
Both options will download the full dataset, including aerial images (as .tif), aerial lidar point clouds (as .copc.laz) and building polygon annotaions (as MS-COCO .json) into `$DATA_ROOT` . The size of the dataset is around 163GB.
### Structure
<details>
<summary>π Click to expand dataset folder structure</summary -->
```text
PixelsPointsPolygons/data/224
βββ annotations
β βββ annotations_all_test.json
β βββ annotations_all_train.json
β βββ annotations_all_val.json
β ... (24 files total)
βββ images
β βββ train
β β βββ CH
β β β βββ 0
β β β β βββ image0_CH_train.tif
β β β β βββ image1000_CH_train.tif
β β β β βββ image1001_CH_train.tif
β β β β ... (5000 files total)
β β β βββ 5000
β β β β βββ image5000_CH_train.tif
β β β β βββ image5001_CH_train.tif
β β β β βββ image5002_CH_train.tif
β β β β ... (5000 files total)
β β β βββ 10000
β β β βββ image10000_CH_train.tif
β β β βββ image10001_CH_train.tif
β β β βββ image10002_CH_train.tif
β β β ... (5000 files total)
β β β ... (11 dirs total)
β β βββ NY
β β β βββ 0
β β β β βββ image0_NY_train.tif
β β β β βββ image1000_NY_train.tif
β β β β βββ image1001_NY_train.tif
β β β β ... (5000 files total)
β β β βββ 5000
β β β β βββ image5000_NY_train.tif
β β β β βββ image5001_NY_train.tif
β β β β βββ image5002_NY_train.tif
β β β β ... (5000 files total)
β β β βββ 10000
β β β βββ image10000_NY_train.tif
β β β βββ image10001_NY_train.tif
β β β βββ image10002_NY_train.tif
β β β ... (5000 files total)
β β β ... (11 dirs total)
β β βββ NZ
β β βββ 0
β β β βββ image0_NZ_train.tif
β β β βββ image1000_NZ_train.tif
β β β βββ image1001_NZ_train.tif
β β β ... (5000 files total)
β β βββ 5000
β β β βββ image5000_NZ_train.tif
β β β βββ image5001_NZ_train.tif
β β β βββ image5002_NZ_train.tif
β β β ... (5000 files total)
β β βββ 10000
β β βββ image10000_NZ_train.tif
β β βββ image10001_NZ_train.tif
β β βββ image10002_NZ_train.tif
β β ... (5000 files total)
β β ... (11 dirs total)
β βββ val
β β βββ CH
β β β βββ 0
β β β βββ image0_CH_val.tif
β β β βββ image100_CH_val.tif
β β β βββ image101_CH_val.tif
β β β ... (529 files total)
β β βββ NY
β β β βββ 0
β β β βββ image0_NY_val.tif
β β β βββ image100_NY_val.tif
β β β βββ image101_NY_val.tif
β β β ... (529 files total)
β β βββ NZ
β β βββ 0
β β βββ image0_NZ_val.tif
β β βββ image100_NZ_val.tif
β β βββ image101_NZ_val.tif
β β ... (529 files total)
β βββ test
β βββ CH
β β βββ 0
β β β βββ image0_CH_test.tif
β β β βββ image1000_CH_test.tif
β β β βββ image1001_CH_test.tif
β β β ... (5000 files total)
β β βββ 5000
β β β βββ image5000_CH_test.tif
β β β βββ image5001_CH_test.tif
β β β βββ image5002_CH_test.tif
β β β ... (5000 files total)
β β βββ 10000
β β βββ image10000_CH_test.tif
β β βββ image10001_CH_test.tif
β β βββ image10002_CH_test.tif
β β ... (4400 files total)
β βββ NY
β β βββ 0
β β β βββ image0_NY_test.tif
β β β βββ image1000_NY_test.tif
β β β βββ image1001_NY_test.tif
β β β ... (5000 files total)
β β βββ 5000
β β β βββ image5000_NY_test.tif
β β β βββ image5001_NY_test.tif
β β β βββ image5002_NY_test.tif
β β β ... (5000 files total)
β β βββ 10000
β β βββ image10000_NY_test.tif
β β βββ image10001_NY_test.tif
β β βββ image10002_NY_test.tif
β β ... (4400 files total)
β βββ NZ
β βββ 0
β β βββ image0_NZ_test.tif
β β βββ image1000_NZ_test.tif
β β βββ image1001_NZ_test.tif
β β ... (5000 files total)
β βββ 5000
β β βββ image5000_NZ_test.tif
β β βββ image5001_NZ_test.tif
β β βββ image5002_NZ_test.tif
β β ... (5000 files total)
β βββ 10000
β βββ image10000_NZ_test.tif
β βββ image10001_NZ_test.tif
β βββ image10002_NZ_test.tif
β ... (4400 files total)
βββ lidar
β βββ train
β β βββ CH
β β β βββ 0
β β β β βββ lidar0_CH_train.copc.laz
β β β β βββ lidar1000_CH_train.copc.laz
β β β β βββ lidar1001_CH_train.copc.laz
β β β β ... (5000 files total)
β β β βββ 5000
β β β β βββ lidar5000_CH_train.copc.laz
β β β β βββ lidar5001_CH_train.copc.laz
β β β β βββ lidar5002_CH_train.copc.laz
β β β β ... (5000 files total)
β β β βββ 10000
β β β βββ lidar10000_CH_train.copc.laz
β β β βββ lidar10001_CH_train.copc.laz
β β β βββ lidar10002_CH_train.copc.laz
β β β ... (5000 files total)
β β β ... (11 dirs total)
β β βββ NY
β β β βββ 0
β β β β βββ lidar0_NY_train.copc.laz
β β β β βββ lidar10_NY_train.copc.laz
β β β β βββ lidar1150_NY_train.copc.laz
β β β β ... (1071 files total)
β β β βββ 5000
β β β β βββ lidar5060_NY_train.copc.laz
β β β β βββ lidar5061_NY_train.copc.laz
β β β β βββ lidar5062_NY_train.copc.laz
β β β β ... (2235 files total)
β β β βββ 10000
β β β βββ lidar10000_NY_train.copc.laz
β β β βββ lidar10001_NY_train.copc.laz
β β β βββ lidar10002_NY_train.copc.laz
β β β ... (4552 files total)
β β β ... (11 dirs total)
β β βββ NZ
β β βββ 0
β β β βββ lidar0_NZ_train.copc.laz
β β β βββ lidar1000_NZ_train.copc.laz
β β β βββ lidar1001_NZ_train.copc.laz
β β β ... (5000 files total)
β β βββ 5000
β β β βββ lidar5000_NZ_train.copc.laz
β β β βββ lidar5001_NZ_train.copc.laz
β β β βββ lidar5002_NZ_train.copc.laz
β β β ... (5000 files total)
β β βββ 10000
β β βββ lidar10000_NZ_train.copc.laz
β β βββ lidar10001_NZ_train.copc.laz
β β βββ lidar10002_NZ_train.copc.laz
β β ... (4999 files total)
β β ... (11 dirs total)
β βββ val
β β βββ CH
β β β βββ 0
β β β βββ lidar0_CH_val.copc.laz
β β β βββ lidar100_CH_val.copc.laz
β β β βββ lidar101_CH_val.copc.laz
β β β ... (529 files total)
β β βββ NY
β β β βββ 0
β β β βββ lidar0_NY_val.copc.laz
β β β βββ lidar100_NY_val.copc.laz
β β β βββ lidar101_NY_val.copc.laz
β β β ... (529 files total)
β β βββ NZ
β β βββ 0
β β βββ lidar0_NZ_val.copc.laz
β β βββ lidar100_NZ_val.copc.laz
β β βββ lidar101_NZ_val.copc.laz
β β ... (529 files total)
β βββ test
β βββ CH
β β βββ 0
β β β βββ lidar0_CH_test.copc.laz
β β β βββ lidar1000_CH_test.copc.laz
β β β βββ lidar1001_CH_test.copc.laz
β β β ... (5000 files total)
β β βββ 5000
β β β βββ lidar5000_CH_test.copc.laz
β β β βββ lidar5001_CH_test.copc.laz
β β β βββ lidar5002_CH_test.copc.laz
β β β ... (5000 files total)
β β βββ 10000
β β βββ lidar10000_CH_test.copc.laz
β β βββ lidar10001_CH_test.copc.laz
β β βββ lidar10002_CH_test.copc.laz
β β ... (4400 files total)
β βββ NY
β β βββ 0
β β β βββ lidar0_NY_test.copc.laz
β β β βββ lidar1000_NY_test.copc.laz
β β β βββ lidar1001_NY_test.copc.laz
β β β ... (4964 files total)
β β βββ 5000
β β β βββ lidar5000_NY_test.copc.laz
β β β βββ lidar5001_NY_test.copc.laz
β β β βββ lidar5002_NY_test.copc.laz
β β β ... (4953 files total)
β β βββ 10000
β β βββ lidar10000_NY_test.copc.laz
β β βββ lidar10001_NY_test.copc.laz
β β βββ lidar10002_NY_test.copc.laz
β β ... (4396 files total)
β βββ NZ
β βββ 0
β β βββ lidar0_NZ_test.copc.laz
β β βββ lidar1000_NZ_test.copc.laz
β β βββ lidar1001_NZ_test.copc.laz
β β ... (5000 files total)
β βββ 5000
β β βββ lidar5000_NZ_test.copc.laz
β β βββ lidar5001_NZ_test.copc.laz
β β βββ lidar5002_NZ_test.copc.laz
β β ... (5000 files total)
β βββ 10000
β βββ lidar10000_NZ_test.copc.laz
β βββ lidar10001_NZ_test.copc.laz
β βββ lidar10002_NZ_test.copc.laz
β ... (4400 files total)
βββ ffl
βββ train
β βββ CH
β β βββ 0
β β β βββ image0_CH_train.pt
β β β βββ image1000_CH_train.pt
β β β βββ image1001_CH_train.pt
β β β ... (5000 files total)
β β βββ 5000
β β β βββ image5000_CH_train.pt
β β β βββ image5001_CH_train.pt
β β β βββ image5002_CH_train.pt
β β β ... (5000 files total)
β β βββ 10000
β β βββ image10000_CH_train.pt
β β βββ image10001_CH_train.pt
β β βββ image10002_CH_train.pt
β β ... (5000 files total)
β β ... (11 dirs total)
β βββ NY
β β βββ 0
β β β βββ image0_NY_train.pt
β β β βββ image1000_NY_train.pt
β β β βββ image1001_NY_train.pt
β β β ... (5000 files total)
β β βββ 5000
β β β βββ image5000_NY_train.pt
β β β βββ image5001_NY_train.pt
β β β βββ image5002_NY_train.pt
β β β ... (5000 files total)
β β βββ 10000
β β βββ image10000_NY_train.pt
β β βββ image10001_NY_train.pt
β β βββ image10002_NY_train.pt
β β ... (5000 files total)
β β ... (11 dirs total)
β βββ NZ
β β βββ 0
β β β βββ image0_NZ_train.pt
β β β βββ image1000_NZ_train.pt
β β β βββ image1001_NZ_train.pt
β β β ... (5000 files total)
β β βββ 5000
β β β βββ image5000_NZ_train.pt
β β β βββ image5001_NZ_train.pt
β β β βββ image5002_NZ_train.pt
β β β ... (5000 files total)
β β βββ 10000
β β βββ image10000_NZ_train.pt
β β βββ image10001_NZ_train.pt
β β βββ image10002_NZ_train.pt
β β ... (5000 files total)
β β ... (11 dirs total)
β βββ processed-flag-all
β βββ processed-flag-CH
β βββ processed-flag-NY
β ... (8 files total)
βββ val
β βββ CH
β β βββ 0
β β βββ image0_CH_val.pt
β β βββ image100_CH_val.pt
β β βββ image101_CH_val.pt
β β ... (529 files total)
β βββ NY
β β βββ 0
β β βββ image0_NY_val.pt
β β βββ image100_NY_val.pt
β β βββ image101_NY_val.pt
β β ... (529 files total)
β βββ NZ
β β βββ 0
β β βββ image0_NZ_val.pt
β β βββ image100_NZ_val.pt
β β βββ image101_NZ_val.pt
β β ... (529 files total)
β βββ processed-flag-all
β βββ processed-flag-CH
β βββ processed-flag-NY
β ... (8 files total)
βββ test
βββ CH
β βββ 0
β β βββ image0_CH_test.pt
β β βββ image1000_CH_test.pt
β β βββ image1001_CH_test.pt
β β ... (5000 files total)
β βββ 5000
β β βββ image5000_CH_test.pt
β β βββ image5001_CH_test.pt
β β βββ image5002_CH_test.pt
β β ... (5000 files total)
β βββ 10000
β βββ image10000_CH_test.pt
β βββ image10001_CH_test.pt
β βββ image10002_CH_test.pt
β ... (4400 files total)
βββ NY
β βββ 0
β β βββ image0_NY_test.pt
β β βββ image1000_NY_test.pt
β β βββ image1001_NY_test.pt
β β ... (5000 files total)
β βββ 5000
β β βββ image5000_NY_test.pt
β β βββ image5001_NY_test.pt
β β βββ image5002_NY_test.pt
β β ... (5000 files total)
β βββ 10000
β βββ image10000_NY_test.pt
β βββ image10001_NY_test.pt
β βββ image10002_NY_test.pt
β ... (4400 files total)
βββ NZ
β βββ 0
β β βββ image0_NZ_test.pt
β β βββ image1000_NZ_test.pt
β β βββ image1001_NZ_test.pt
β β ... (5000 files total)
β βββ 5000
β β βββ image5000_NZ_test.pt
β β βββ image5001_NZ_test.pt
β β βββ image5002_NZ_test.pt
β β ... (5000 files total)
β βββ 10000
β βββ image10000_NZ_test.pt
β βββ image10001_NZ_test.pt
β βββ image10002_NZ_test.pt
β ... (4400 files total)
βββ processed-flag-all
βββ processed-flag-CH
βββ processed-flag-NY
... (8 files total)
```
</details>
## Pretrained model weights
### Download
The recommended and fastest way to download the pretrained model weights is to run
```
python scripts/download_pretrained.py --model-root $MODEL_ROOT
```
Optionally you can also download the weights by running
```
git clone https://huggingface.co/rsi/PixelsPointsPolygons $MODEL_ROOT
```
Both options will download all checkpoints (as .pth) and results presented in the paper (as MS-COCO .json) into `$MODEL_ROOT` .
## Code
### Download
```
git clone https://github.com/raphaelsulzer/PixelsPointsPolygons
```
### Installation
To create a conda environment named `p3` and install the repository as a python package with all dependencies run
```
bash install.sh
```
or, if you want to manage the environment yourself run
```
pip install -r requirements-torch-cuda.txt
pip install .
```
β οΈ **Warning**: The implementation of the LiDAR point cloud encoder uses Open3D-ML. Currently, Open3D-ML officially only supports the PyTorch version specified in `requirements-torch-cuda.txt`.
<!-- ## Model Zoo
| Model | \<model> | Encoder | \<encoder> |Image |LiDAR | IoU | C-IoU |
|--------------- |---- |--------------- |--------------- |--- |--- |----- |----- |
| Frame Field Learning |\<ffl> | Vision Transformer (ViT) | \<vit_cnn> | β
| | 0.85 | 0.90 |
| Frame Field Learning |\<ffl> | PointPillars (PP) + ViT | \<pp_vit_cnn> | | β
| 0.80 | 0.88 |
| Frame Field Learning |\<ffl> | PP+ViT \& ViT | \<fusion_vit_cnn> | β
|β
| 0.78 | 0.85 |
| HiSup |\<hisup> | Vision Transformer (ViT) | \<vit_cnn> | β
| | 0.85 | 0.90 |
| HiSup |\<hisup> | PointPillars (PP) + ViT | \<pp_vit_cnn> | | β
| 0.80 | 0.88 |
| HiSup |\<hisup> | PP+ViT \& ViT | \<fusion_vit> | β
|β
| 0.78 | 0.85 |
| Pix2Poly |\<pix2poly>| Vision Transformer (ViT) | \<vit> | β
| | 0.85 | 0.90 |
| Pix2Poly |\<pix2poly>| PointPillars (PP) + ViT | \<pp_vit> | | β
| 0.80 | 0.88 |
| Pix2Poly |\<pix2poly>| PP+ViT \& ViT | \<fusion_vit> | β
|β
| 0.78 | 0.85 | -->
### Setup
The project supports hydra configuration which allows to modify any parameter either from a `.yaml` file or directly from the command line.
To setup the project structure we recommend to specify your `$DATA_ROOT` and `$MODEL_ROOT` in `config/host/default.yaml`.
To view all available configuration options run
```
python scripts/train.py --help
```
<!-- The most important parameters are described below:
<details>
<summary>CLI Parameters</summary>
```text
βββ processed-flag-all
βββ processed-flag-CH
βββ processed-flag-NY
... (8 files total)
```
</details> -->
### Predict demo tile
After downloading the model weights and setting up the code you can predict a demo tile by running
```
python scripts/predict_demo.py checkpoint=best_val_iou experiment=$MODEL_$MODALITY +image_file=demo_data/image0_CH_val.tif +lidar_file=demo_data/lidar0_CH_val.copc.laz
```
At least one of `image_file` or `lidar_file` has to be specified. `$MODEL` can be one of the following: `ffl`, `hisup` or `p2p`. `$MODALITY` can be `image`, `lidar` or `fusion`.
The result will be stored in `prediction.png`.
### Reproduce paper results
To reproduce the results from the paper you can run the following commands
```
python scripts/modality_ablation.py
python scripts/lidar_density_ablation.py
python scripts/all_countries.py
```
### Custom training, prediction and evaluation
We recommend to first setup a custom experiment file `$EXP_FILE` in `config/experiment/` following the structure of one of the existing files, e.g. `ffl_fusion.yaml`. You can then run
```
# train your model (on multiple GPUs)
torchrun --nproc_per_node=$NUM_GPU scripts/train.py experiment=$EXP_FILE
# predict the test set with your model (on multiple GPUs)
torchrun --nproc_per_node=$NUM_GPU scripts/predict.py experiment=$EXP_FILE evaluation=test checkpoint=best_val_iou
# evaluate your prediction of the test set
python scripts/evaluate.py experiment=$EXP_FILE evaluation=test checkpoint=best_val_iou
```
You could also continue training from a provided pretrained model with
```
# train your model (on a single GPU)
python scripts/train.py experiment=p2p_fusion checkpoint=latest
```
## Citation
If you use our work please cite
```bibtex
@misc{sulzer2025p3datasetpixelspoints,
title={The P$^3$ dataset: Pixels, Points and Polygons for Multimodal Building Vectorization},
author={Raphael Sulzer and Liuyun Duan and Nicolas Girard and Florent Lafarge},
year={2025},
eprint={2505.15379},
archivePrefix={arXiv},
primaryClass={cs.CV},
url={https://arxiv.org/abs/2505.15379},
}
```
## Acknowledgements
This repository benefits from the following open-source work. We thank the authors for their great work.
1. [Frame Field Learning](https://github.com/Lydorn/Polygonization-by-Frame-Field-Learning)
2. [HiSup](https://github.com/SarahwXU/HiSup)
3. [Pix2Poly](https://github.com/yeshwanth95/Pix2Poly) |