CV-DoG8k-HardNet-LightGlue

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: Challenge organizers (contact)
• Keypoint: cv2dog
• Descriptor: hardnet (128 float32: 512 bytes)
• Number of features: 8000
• Summary: OpenCV SIFT keypoints 8000 features, with HardNet descriptor as implemented in kornia. Matching with LightGlue, trained by kornia team, and DEGENSAC
• Paper: https://arxiv.org/pdf/2306.13643.pdf
• Website: https://github.com/kornia/kornia
• Processing date: 24-01-24

Phototourism dataset / Stereo track

{
    "use_custom_matches": true,
    "custom_matches_name": "lg3",
    "geom": {
        "method": "cmp-degensac-f",
        "threshold": 0.5,
        "confidence": 0.999999,
        "max_iter": 100000,
        "error_type": "sampson",
        "degeneracy_check": true
    }
}

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

{
    "use_custom_matches": true,
    "custom_matches_name": "lg3",
    "colmap": {}
}

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

{
    "use_custom_matches": true,
    "custom_matches_name": "lg3",
    "geom": {
        "method": "cmp-degensac-f",
        "threshold": 1.5,
        "confidence": 0.999999,
        "max_iter": 100000,
        "error_type": "sampson",
        "degeneracy_check": true
    }
}

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.

---

— Lizard —

— Pond —

— Tree —

Prague Parks dataset / Multiview track

{
    "use_custom_matches": true,
    "custom_matches_name": "lg3",
    "colmap": {}
}

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.

---

— Lizard —

— Pond —

— Tree —

Google Urban dataset / Stereo track

Google Urban dataset / Multiview track