Added DoG-HardNet extractor and configs

README.md

@@ -157,13 +157,16 @@ python -m gluefactory.eval.megadepth1500 --conf gluefactory/configs/superpoint+l
 
 Here are the results as Area Under the Curve (AUC) of the pose error at  5/10/20 degrees:
 
-| Methods                                                      | [pycolmap](../gluefactory/robust_estimators/relative_pose/pycolmap.py)         | [OpenCV](../gluefactory/robust_estimators/relative_pose/opencv.py)       | [PoseLib](../gluefactory/robust_estimators/relative_pose/poselib.py)      |
+| Methods (2K if not specified)                                                      | [pycolmap](../gluefactory/robust_estimators/relative_pose/pycolmap.py)         | [OpenCV](../gluefactory/robust_estimators/relative_pose/opencv.py)       | [PoseLib](../gluefactory/robust_estimators/relative_pose/poselib.py)      |
 | ------------------------------------------------------------ | ------------------ | ------------------ | ------------------ |
 | [SuperPoint + SuperGlue](gluefactory/configs/superpoint+superglue-official.yaml) | 54.4 / 70.4 / 82.4 | 48.7 / 65.6 / 79.0 | 64.8 / 77.9 / 87.0 |
-| [SuperPoint + LightGlue](gluefactory/configs/superpoint+lightglue-official.yaml) | 56.7 / 72.4 / 83.7 | 51.0 / 68.1 / 80.7 | 66.8 / 79.3 / 87.9 |
-| [SIFT (2K) + LightGlue](gluefactory/configs/sift+lightglue-official.yaml) | ? / ? / ? | 43.5 / 61.5 / 75.9 | 60.4 / 74.3 / 84.5 |
-| [SIFT (4K) + LightGlue](gluefactory/configs/sift+lightglue-official.yaml) | ? / ? / ? | 49.9 / 67.3 / 80.3 | 65.9 / 78.6 / 87.4 |
-| [ALIKED + LightGlue](gluefactory/configs/aliked+lightglue-official.yaml) | ? / ? / ? | 51.5 / 68.1 / 80.4 | 66.3 / 78.7 / 87.5 |
+| [SuperPoint + LightGlue](gluefactory/configs/superpoint+lightglue-official.yaml) | 56.7 / 72.4 / 83.7 | 51.0 / 68.1 / 80.7 | **66.8** / **79.3** / **87.9** |
+| [ALIKED + LightGlue](gluefactory/configs/aliked+lightglue-official.yaml) | **59.6** / **74.1** / **84.6** | 51.5 / 68.1 / 80.4 | 66.3 / 78.7 / 87.5 |
+| [DISK + LightGlue](gluefactory/configs/disk+lightglue-official.yaml) | 53.6 / 69.2 / 80.7  |  45.1 / 62.7 / 76.6  |  61.3 / 74.3 / 83.8  |
+| [SIFT + LightGlue](gluefactory/configs/sift+lightglue-official.yaml) | 51.1 / 67.1 / 78.9 | 43.5 / 61.5 / 75.9 | 60.4 / 74.3 / 84.5 |
+| [SIFT (4K) + LightGlue](gluefactory/configs/sift+lightglue-official.yaml) | 56.2 / 71.8 / 83.3  | 49.9 / 67.3 / 80.3 | 65.9 / 78.6 / 87.4 |
+| [DoGHardNet + LightGlue](gluefactory/configs/doghardnet+lightglue-official.yaml) | 53.1 / 69.6 / 81.5 | 49.8 / 66.2 / 79.0 | 61.0 / 74.9 / 84.7 |
+| [DoGHardNet (4K) + LightGlue](gluefactory/configs/doghardnet+lightglue-official.yaml) | 58.8 / 73.6 / 84.4 | **53.4** / **69.7** / **81.7**  | 66.5 / 79.0 / 87.6 |
 | [SuperPoint + GlueStick](gluefactory/configs/superpoint+lsd+gluestick.yaml) | 53.2 / 69.8 / 81.9 | 46.3 / 64.2 / 78.1 | 64.4 / 77.5 / 86.5 |
 
 </details>
@@ -304,6 +307,7 @@ Using the following local feature extractors:
 | SIFT (via [pycolmap](https://github.com/colmap/pycolmap)) | `sift+lightglue_{homography,megadepth}.yaml` |
 | [ALIKED](https://github.com/Shiaoming/ALIKED) | `aliked+lightglue_{homography,megadepth}.yaml` |
 | [DISK](https://github.com/cvlab-epfl/disk) | `disk+lightglue_{homography,megadepth}.yaml` |
+| [DoG-HardNet](https://github.com/DagnyT/hardnet/) | `doghardnet+lightglue_{homography,megadepth}.yaml` |
 | Key.Net + HardNet | ❌ TODO |
 
 ## Coming soon

gluefactory/configs/doghardnet+lightglue-official.yaml

@@ -0,0 +1,28 @@
+model:
+    name: two_view_pipeline
+    extractor:
+      name: extractors.doghardnet
+      backend: opencv
+      max_num_keypoints: 2048
+    matcher:
+      name: matchers.lightglue_pretrained
+      features: doghardnet
+      depth_confidence: -1
+      width_confidence: -1
+      filter_threshold: 0.1
+benchmarks:
+    megadepth1500:
+      data:
+        preprocessing:
+          side: long
+          resize: 1600
+      eval:
+        estimator: opencv
+        ransac_th: 0.5
+    hpatches:
+      eval:
+        estimator: opencv
+        ransac_th: 0.5
+      model:
+        extractor:
+          max_num_keypoints: 1024  # overwrite config above

gluefactory/configs/doghardnet+lightglue_homography.yaml

@@ -0,0 +1,52 @@
+data:
+    name: homographies
+    data_dir: revisitop1m
+    train_size: 150000
+    val_size: 2000
+    batch_size: 64
+    num_workers: 14
+    homography:
+        difficulty: 0.7
+        max_angle: 45
+    photometric:
+        name: lg
+model:
+    name: two_view_pipeline
+    extractor:
+        name: extractors.doghardnet
+        backend: opencv
+        max_num_keypoints: 1024
+        force_num_keypoints: true
+        nms_radius: 3
+        trainable: false
+    ground_truth:
+        name: matchers.homography_matcher
+        th_positive: 3
+        th_negative: 3
+    matcher:
+        name: matchers.lightglue
+        filter_threshold: 0.1
+        flash: false
+        add_scale_ori: true
+        checkpointed: true
+        input_dim: 128
+train:
+    seed: 0
+    epochs: 40
+    log_every_iter: 100
+    eval_every_iter: 500
+    lr: 1e-4
+    lr_schedule:
+        start: 20
+        type: exp
+        on_epoch: true
+        exp_div_10: 10
+    plot: [5, 'gluefactory.visualization.visualize_batch.make_match_figures']
+benchmarks:
+    hpatches:
+      eval:
+        estimator: opencv
+        ransac_th: 0.5
+      model:
+        extractor:
+          nms_radius: 0

gluefactory/configs/doghardnet+lightglue_megadepth.yaml

@@ -0,0 +1,79 @@
+data:
+    name: megadepth
+    preprocessing:
+        resize: 1024
+        side: long
+        square_pad: True
+    train_split: train_scenes_clean.txt
+    train_num_per_scene: 300
+    val_split: valid_scenes_clean.txt
+    val_pairs: valid_pairs.txt
+    min_overlap: 0.1
+    max_overlap: 0.7
+    num_overlap_bins: 3
+    read_depth: true
+    read_image: true
+    batch_size: 16
+    num_workers: 14
+    load_features:
+        do: false  # enable this if you have cached predictions
+        path: exports/megadepth-undist-depth-r1024_doghardnet-k4096/{scene}.h5
+        padding_length: 4096
+        padding_fn: pad_local_features
+        data_keys: ["keypoints", "keypoint_scores", "descriptors", "oris", "scales"]
+model:
+    name: two_view_pipeline
+    extractor:
+        name: extractors.doghardnet
+        backend: opencv
+        max_num_keypoints: 4096
+        force_num_keypoints: True
+        nms_radius: 3
+        trainable: False
+    matcher:
+        name: matchers.lightglue
+        filter_threshold: 0.1
+        flash: false
+        checkpointed: true
+        add_scale_ori: true
+        input_dim: 128
+    ground_truth:
+        name: matchers.depth_matcher
+        th_positive: 3
+        th_negative: 5
+        th_epi: 5
+    allow_no_extract: True
+train:
+    seed: 10
+    epochs: 50
+    log_every_iter: 100
+    eval_every_iter: 1000
+    lr: 1e-4
+    lr_schedule:
+        start: 30
+        type: exp
+        on_epoch: true
+        exp_div_10: 10
+    dataset_callback_fn: sample_new_items
+    plot: [5, 'gluefactory.visualization.visualize_batch.make_match_figures']
+benchmarks:
+    megadepth1500:
+        data:
+            preprocessing:
+                side: long
+                resize: 1600
+        model:
+            extractor:
+                nms_radius: 0
+                max_num_keypoints: 2048
+        eval:
+            estimator: opencv
+            ransac_th: 0.5
+    hpatches:
+        eval:
+            estimator: opencv
+            ransac_th: 0.5
+        model:
+            extractor:
+                max_num_keypoints: 1024
+                nms_radius: 0