Submission ID: 7ab671c5
PoSFeat
Processed: 21-11-12. Download link: 7ab671c5a8d829ef-posfeat-2k-both-degensac-1103.json
This page ranks the submission against all others using the same number of keypoints, regardless of descriptor size. Please hover over table headers for descriptions on metrics and full scene names.
- Phototourism dataset: Stereo track / Multiview track
- Prague Parks dataset: Stereo track / Multiview track
- Google Urban dataset: Stereo track / Multiview track
Metadata
- Authors: Kunhong Li, Longguang Wang and Yulan Guo (contact)
- Keypoint: posfeat2k-1103
- Descriptor: posfeat2k-1103 (128 float32: 512 bytes)
- Number of features: 2048
- Summary: PosFeat with 2048 features, using the built-in matcher (bidirectional filter with the both strategy, optimal inlier and ratio test thresholds) with DEGENSAC.
- Paper: N/A
- Website: N/A
- Processing date: 21-11-12
Phototourism dataset / Stereo track
mAA at 10 degrees: 0.46237 (±0.00000 over 1 run(s) / ±0.12993 over 9 scenes)
Rank (per category): 54 (of 87)
| Scene | Features | Matches (raw) |
Matches (final) |
Rep. @ 3 px. | MS @ 3 px. | mAA(5o) | mAA(10o) |
| BM | 2048.0 | 548.4 | 467.0 | 0.592 Rank: 11/87 |
0.934 Rank: 17/87 |
0.21855 (±0.00000) Rank: 54/87 |
0.34620 (±0.00000) Rank: 54/87 |
| FCS | 2048.0 | 458.0 | 357.1 | 0.389 Rank: 53/87 |
0.863 Rank: 47/87 |
0.50811 (±0.00000) Rank: 58/87 |
0.63708 (±0.00000) Rank: 59/87 |
| LMS | 2048.0 | 424.9 | 355.5 | 0.467 Rank: 13/87 |
0.704 Rank: 11/87 |
0.48414 (±0.00000) Rank: 57/87 |
0.60381 (±0.00000) Rank: 57/87 |
| LB | 2048.0 | 374.1 | 303.0 | 0.465 Rank: 6/87 |
0.732 Rank: 18/87 |
0.41140 (±0.00000) Rank: 54/87 |
0.51436 (±0.00000) Rank: 51/87 |
| MC | 2048.0 | 557.4 | 466.9 | 0.491 Rank: 22/87 |
0.914 Rank: 22/87 |
0.30668 (±0.00000) Rank: 54/87 |
0.46211 (±0.00000) Rank: 47/87 |
| MR | 2048.0 | 431.3 | 349.7 | 0.387 Rank: 61/87 |
0.906 Rank: 20/87 |
0.22937 (±0.00000) Rank: 46/87 |
0.32332 (±0.00000) Rank: 52/87 |
| PSM | 2048.0 | 258.8 | 200.7 | 0.344 Rank: 35/87 |
0.607 Rank: 38/87 |
0.11525 (±0.00000) Rank: 57/87 |
0.22381 (±0.00000) Rank: 55/87 |
| SF | 2048.0 | 431.9 | 333.4 | 0.344 Rank: 69/87 |
0.784 Rank: 42/87 |
0.42297 (±0.00000) Rank: 51/87 |
0.54810 (±0.00000) Rank: 55/87 |
| SPC | 2048.0 | 386.8 | 301.4 | 0.427 Rank: 21/87 |
0.812 Rank: 21/87 |
0.35323 (±0.00000) Rank: 56/87 |
0.50254 (±0.00000) Rank: 56/87 |
| Avg | 2048.0 | 430.2 | 348.3 | 0.434 Rank: 16/87 |
0.806 Rank: 15/87 |
0.33885 (±0.00000) Rank: 54/87 |
0.46237 (±0.00000) Rank: 54/87 |
We show the inliers that survive the robust estimation loop (i.e. RANSAC), or those supplied with the submission if using custom matches, and use the depth estimates to determine whether they are correct. We draw matches above a 5-pixel error threshold in red, and those below are color-coded by their error, from 0 (green) to 5 pixels (yellow). Matches for which we do not have depth estimates are drawn in blue. Please note that the depth maps are estimates and may contain errors.
— British Museum —
— Florence Cathedral Side —
— Lincoln Memorial Statue —
— London Bridge —
— Milan Cathedral —
— Mount Rushmore —
— Piazza San Marco —
— Sagrada Familia —
— Saint Paul's Cathedral —
Phototourism dataset / Multiview track
mAA at 10 degrees: N/A (±0.00000 over 1 run(s) / ±0.10967 over 9 scenes)
Rank (per category): 53 (of 87)
| Scene | Features | Matches (input) |
RegistrationRatio (%) | Number of Landmarks |
Track Length | ATE | mAA(50) | mAA(100) |
| BM | 2048.0 | 548.65 | 100.00 Rank: 1/87 |
2059.49 Rank: 33/87 |
6.085 Rank: 11/87 |
0.43217 Rank: 47/87 |
0.51231 (±0.00000) Rank: 49/87 |
0.65474 (±0.00000) Rank: 47/87 |
| FCS | 2048.0 | 476.03 | 96.15 Rank: 69/87 |
2541.11 Rank: 36/87 |
4.819 Rank: 53/87 |
0.26362 Rank: 52/87 |
0.69103 (±0.00000) Rank: 63/87 |
0.75850 (±0.00000) Rank: 65/87 |
| LMS | 2048.0 | 460.05 | 99.20 Rank: 42/87 |
2029.15 Rank: 23/87 |
5.519 Rank: 39/87 |
0.30288 Rank: 47/87 |
0.81138 (±0.00000) Rank: 54/87 |
0.86655 (±0.00000) Rank: 55/87 |
| LB | 2048.0 | 445.30 | 97.56 Rank: 62/87 |
1930.78 Rank: 38/87 |
5.657 Rank: 14/87 |
0.46699 Rank: 9/87 |
0.68686 (±0.00000) Rank: 47/87 |
0.77845 (±0.00000) Rank: 48/87 |
| MC | 2048.0 | 546.95 | 100.00 Rank: 1/87 |
2359.86 Rank: 25/87 |
5.702 Rank: 13/87 |
0.41346 Rank: 61/87 |
0.51845 (±0.00000) Rank: 47/87 |
0.66558 (±0.00000) Rank: 48/87 |
| MR | 2048.0 | 424.38 | 93.40 Rank: 53/87 |
2435.05 Rank: 19/87 |
4.604 Rank: 59/87 |
0.59586 Rank: 70/87 |
0.39807 (±0.00000) Rank: 43/87 |
0.51539 (±0.00000) Rank: 43/87 |
| PSM | 2048.0 | 256.82 | 94.62 Rank: 55/87 |
2508.08 Rank: 31/87 |
3.763 Rank: 57/87 |
0.54752 Rank: 54/87 |
0.46741 (±0.00000) Rank: 52/87 |
0.55731 (±0.00000) Rank: 52/87 |
| SF | 2048.0 | 423.73 | 96.40 Rank: 61/87 |
2732.98 Rank: 32/87 |
4.595 Rank: 62/87 |
0.29738 Rank: 44/87 |
0.70318 (±0.00000) Rank: 57/87 |
0.78418 (±0.00000) Rank: 58/87 |
| SPC | 2048.0 | 401.49 | 99.07 Rank: 60/87 |
2207.11 Rank: 40/87 |
5.236 Rank: 16/87 |
0.56710 Rank: 66/87 |
0.67453 (±0.00000) Rank: 61/87 |
0.78126 (±0.00000) Rank: 62/87 |
| Avg | 2048.0 | 442.60 | 97.38 Rank: 58/87 |
2311.51 Rank: 32/87 |
5.109 Rank: 30/87 |
0.43189 Rank: 58/87 |
0.60703 (±0.00000) Rank: 50/87 |
0.70688 (±0.00000) Rank: 53/87 |
In the multi-view track we reconstruct the scene with Structure-from-Motion (Colmap) with small sets of images. We show the results for one bag of 10 images. Keypoints are drawn in blue if they are part of the model, and in red otherwise.
— British Museum —
— Florence Cathedral Side —
— Lincoln Memorial Statue —
— London Bridge —
— Milan Cathedral —
— Mount Rushmore —
— Piazza San Marco —
— Sagrada Familia —
— Saint Paul's Cathedral —
Prague Parks dataset / Stereo track
mAA at 10 degrees: N/A (±N/A over N/A run(s) / ±N/A over 3 scenes)
Rank (per category): (of 86)
No results for this dataset.
Prague Parks dataset / Multiview track
mAA at 10 degrees: N/A (±N/A over N/A run(s) / ±N/A over 3 scenes)
Rank (per category): (of 86)
No results for this dataset.
Google Urban dataset / Stereo track
mAA at 10 degrees: N/A (±N/A over N/A run(s) / ±N/A over 17 scenes)
Rank (per category): (of 86)
No results for this dataset.
Google Urban dataset / Multiview track
mAA at 10 degrees: N/A (±N/A over N/A run(s) / ±N/A over 17 scenes)
Rank (per category): (of 86)
No results for this dataset.