Submission ID: 00104
VL-DoG-SIFT-8k
Processed: 20-05-04. Download link: sid-00104-vl-dog-sift-8k-both-ransac.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.
Metadata
- Authors: Challenge organizers (contact)
- Keypoint: dog
- Descriptor: vlsift (128 float32: 512 bytes)
- Number of features: 8000
- Summary: VL-DoG-SIFT with 8000 features, using the built-in matcher (bidirectional filter with the both strategy, optimal inlier and ratio test thresholds) with RANSAC
- Paper: N/A
- Website: N/A
- Origin: Baseline
- Flags: is_baseline
Phototourism / Stereo track
mAA at 10 degrees: 0.39987 (±0.00068 over 3 run(s) / ±0.13291 over 9 scenes)
Rank (per category): 120 (of 147)
| Scene | Features | Matches (matcher) |
Matches (filter) |
Matches (final) |
Rep. @ 3 px. | MS @ 3 px. | mAA(5o) | mAA(10o) |
| bm | 7973.7 | 439.2 | 439.2 | 198.6 | 0.562 Rank: 17/147 |
0.954 Rank: 17/147 |
0.09856 (±0.00175) Rank: 124/147 |
0.18733 (±0.00248) Rank: 125/147 |
| fcs | 7906.0 | 459.1 | 459.1 | 221.7 | 0.448 Rank: 75/147 |
0.943 Rank: 22/147 |
0.46924 (±0.00246) Rank: 121/147 |
0.60829 (±0.00217) Rank: 121/147 |
| lms | 7802.8 | 368.1 | 368.1 | 132.4 | 0.503 Rank: 17/147 |
0.717 Rank: 10/147 |
0.37176 (±0.00244) Rank: 123/147 |
0.46510 (±0.00108) Rank: 124/147 |
| lb | 7963.2 | 335.5 | 335.5 | 131.9 | 0.487 Rank: 73/147 |
0.745 Rank: 82/147 |
0.28446 (±0.00175) Rank: 124/147 |
0.38304 (±0.00188) Rank: 124/147 |
| mc | 7845.1 | 467.1 | 467.1 | 213.8 | 0.534 Rank: 74/147 |
0.968 Rank: 10/147 |
0.30445 (±0.00175) Rank: 124/147 |
0.45424 (±0.00593) Rank: 115/147 |
| mr | 7801.8 | 465.1 | 465.1 | 244.8 | 0.470 Rank: 76/147 |
0.975 Rank: 35/147 |
0.23863 (±0.00670) Rank: 117/147 |
0.33367 (±0.00665) Rank: 117/147 |
| psm | 7834.1 | 217.3 | 217.3 | 75.9 | 0.394 Rank: 76/147 |
0.474 Rank: 119/147 |
0.10753 (±0.00297) Rank: 118/147 |
0.20002 (±0.00292) Rank: 115/147 |
| sf | 7917.0 | 579.2 | 579.2 | 257.6 | 0.498 Rank: 73/147 |
0.873 Rank: 73/147 |
0.38991 (±0.00121) Rank: 121/147 |
0.53128 (±0.00068) Rank: 123/147 |
| spc | 7881.6 | 368.6 | 368.6 | 140.5 | 0.511 Rank: 17/147 |
0.865 Rank: 58/147 |
0.29132 (±0.00267) Rank: 121/147 |
0.43585 (±0.00273) Rank: 121/147 |
| avg | 7880.6 | 411.0 | 411.0 | 179.7 | 0.490 Rank: 20/147 |
0.835 Rank: 80/147 |
0.28398 (±0.00030) Rank: 119/147 |
0.39987 (±0.00068) Rank: 120/147 |
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 / Multiview track
mAA at 10 degrees: 0.62829 (±0.00112 over 3 run(s) / ±0.17045 over 9 scenes)
Rank (per category): 116 (of 147)
| Scene | Features | Matches (input) |
RegistrationRatio (%) | Number of Landmarks |
Track Length | ATE | mAA(50) | mAA(100) |
| bm | 7973.7 | 334.21 | 99.69 Rank: 89/147 |
2362.70 Rank: 111/147 |
4.894 Rank: 87/147 |
0.60244 Rank: 115/147 |
0.33760 (±0.00529) Rank: 117/147 |
0.48262 (±0.00366) Rank: 116/147 |
| fcs | 7906.0 | 380.64 | 97.38 Rank: 103/147 |
3467.35 Rank: 122/147 |
4.346 Rank: 101/147 |
0.27046 Rank: 108/147 |
0.68900 (±0.00415) Rank: 115/147 |
0.76130 (±0.00573) Rank: 114/147 |
| lms | 7802.8 | 299.89 | 98.50 Rank: 111/147 |
2573.53 Rank: 128/147 |
4.370 Rank: 98/147 |
0.31731 Rank: 106/147 |
0.78017 (±0.00768) Rank: 112/147 |
0.84095 (±0.00745) Rank: 112/147 |
| lb | 7963.2 | 304.17 | 96.32 Rank: 130/147 |
2492.28 Rank: 126/147 |
4.350 Rank: 107/147 |
0.52590 Rank: 104/147 |
0.52298 (±0.00738) Rank: 118/147 |
0.62151 (±0.00499) Rank: 121/147 |
| mc | 7845.1 | 352.48 | 99.52 Rank: 89/147 |
3199.38 Rank: 113/147 |
4.423 Rank: 105/147 |
0.41989 Rank: 100/147 |
0.49494 (±0.00755) Rank: 86/147 |
0.64113 (±0.00819) Rank: 90/147 |
| mr | 7801.8 | 384.03 | 94.42 Rank: 58/147 |
3795.73 Rank: 118/147 |
4.043 Rank: 92/147 |
0.48104 Rank: 38/147 |
0.40068 (±0.00781) Rank: 102/147 |
0.51669 (±0.01166) Rank: 99/147 |
| psm | 7834.1 | 132.71 | 77.73 Rank: 131/147 |
1833.15 Rank: 134/147 |
2.843 Rank: 125/147 |
0.76243 Rank: 66/147 |
0.18825 (±0.00242) Rank: 130/147 |
0.26989 (±0.00322) Rank: 131/147 |
| sf | 7917.0 | 462.17 | 97.04 Rank: 104/147 |
4616.45 Rank: 117/147 |
4.260 Rank: 103/147 |
0.35606 Rank: 117/147 |
0.69789 (±0.00255) Rank: 103/147 |
0.78660 (±0.00153) Rank: 97/147 |
| spc | 7881.6 | 271.28 | 98.22 Rank: 112/147 |
2935.46 Rank: 123/147 |
4.026 Rank: 115/147 |
0.46669 Rank: 54/147 |
0.63443 (±0.00437) Rank: 111/147 |
0.73394 (±0.00351) Rank: 111/147 |
| avg | 7880.6 | 324.62 | 95.42 Rank: 126/147 |
3030.67 Rank: 123/147 |
4.173 Rank: 109/147 |
0.46691 Rank: 111/147 |
0.52733 (±0.00023) Rank: 112/147 |
0.62829 (±0.00112) Rank: 116/147 |
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 25 images (displaying: 10). 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 —
