limap.point2d.superpoint package

Submodules

limap.point2d.superpoint.main module

limap.point2d.superpoint.main.map_tensor(input_, func)

limap.point2d.superpoint.main.run_superpoint(conf: Dict, image_dir: Path, export_dir: Path | None = None, as_half: bool = True, image_list: Path | List[str] | None = None, feature_path: Path | None = None, overwrite: bool = False, keypoints=None) → Path

limap.point2d.superpoint.superpoint module

class limap.point2d.superpoint.superpoint.SuperPoint(config)

Bases: Module

SuperPoint Convolutional Detector and Descriptor

SuperPoint: Self-Supervised Interest Point Detection and Description. Daniel DeTone, Tomasz Malisiewicz, and Andrew Rabinovich. In CVPRW, 2019. https://arxiv.org/abs/1712.07629

compute_dense_descriptor(data)

Compute keypoints, scores, descriptors for image

compute_dense_descriptor_and_score(data)

Compute dense scores and descriptors for an image

default_config = {'descriptor_dim': 256, 'keypoint_threshold': 0.005, 'max_keypoints': -1, 'nms_radius': 4, 'remove_borders': 4, 'weight_path': None}

download_model(path)

forward(data)

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

sample_descriptors(data, keypoints)

training: bool

limap.point2d.superpoint.superpoint.remove_borders(keypoints, scores, border: int, height: int, width: int)

Removes keypoints too close to the border

limap.point2d.superpoint.superpoint.sample_descriptors(keypoints, descriptors, s: int = 8)

Interpolate descriptors at keypoint locations

limap.point2d.superpoint.superpoint.simple_nms(scores, nms_radius: int)

Fast Non-maximum suppression to remove nearby points

limap.point2d.superpoint.superpoint.top_k_keypoints(keypoints, scores, k: int)

Module contents