limap.line2d.SOLD2 package

Subpackages

• limap.line2d.SOLD2.misc package
  • Submodules
  • limap.line2d.SOLD2.misc.geometry_utils module
    • get_common_line_mask()
    • get_dist_mask()
    • keypoints_to_grid()
    • mask_lines()
    • mask_points()
    • sample_line_points()
    • warp_points()
  • limap.line2d.SOLD2.misc.train_utils module
    • convert_image()
    • get_latest_checkpoint()
    • parse_h5_data()
    • remove_old_checkpoints()
  • limap.line2d.SOLD2.misc.visualize_util module
    • plot_color_line_matches()
    • plot_color_lines()
    • plot_images()
    • plot_junctions()
    • plot_keypoints()
    • plot_line_matches()
    • plot_line_segments()
    • plot_line_segments_from_segments()
    • plot_lines()
    • plot_matches()
    • plot_subsegment_matches()
  • Module contents
• limap.line2d.SOLD2.model package
  • Subpackages
    • limap.line2d.SOLD2.model.nets package
      • Submodules
      • limap.line2d.SOLD2.model.nets.backbone module
      • limap.line2d.SOLD2.model.nets.descriptor_decoder module
      • limap.line2d.SOLD2.model.nets.heatmap_decoder module
      • limap.line2d.SOLD2.model.nets.junction_decoder module
      • limap.line2d.SOLD2.model.nets.lcnn_hourglass module
      • Module contents
  • Submodules
  • limap.line2d.SOLD2.model.line_detection module
    • LineSegmentDetectionModule
      • LineSegmentDetectionModule.candidate_suppression()
      • LineSegmentDetectionModule.convert_inputs()
      • LineSegmentDetectionModule.detect()
      • LineSegmentDetectionModule.detect_bilinear()
      • LineSegmentDetectionModule.detect_local_max()
      • LineSegmentDetectionModule.refine_heatmap()
      • LineSegmentDetectionModule.refine_heatmap_local()
      • LineSegmentDetectionModule.refine_junction_perturb()
      • LineSegmentDetectionModule.segments_to_line_map()
  • limap.line2d.SOLD2.model.line_detector module
    • LineDetector
    • line_map_to_segments()
  • limap.line2d.SOLD2.model.line_matcher module
    • LineMatcher
      • LineMatcher.line_detection()
      • LineMatcher.multiscale_line_detection()
  • limap.line2d.SOLD2.model.line_matching module
    • WunschLineMatcher
      • WunschLineMatcher.asl_feat_saliency_score()
      • WunschLineMatcher.compute_descriptors()
      • WunschLineMatcher.compute_matches()
      • WunschLineMatcher.compute_matches_topk()
      • WunschLineMatcher.compute_matches_topk_gpu()
      • WunschLineMatcher.d2_net_saliency_score()
      • WunschLineMatcher.filter_and_match_lines()
      • WunschLineMatcher.filter_and_match_lines_topk()
      • WunschLineMatcher.filter_and_match_lines_topk_gpu()
      • WunschLineMatcher.forward()
      • WunschLineMatcher.get_pairwise_distance()
      • WunschLineMatcher.needleman_wunsch()
      • WunschLineMatcher.sample_line_points()
      • WunschLineMatcher.sample_salient_points()
  • limap.line2d.SOLD2.model.loss module
    • HeatmapLoss
      • HeatmapLoss.forward()
      • HeatmapLoss.training
    • JunctionDetectionLoss
      • JunctionDetectionLoss.forward()
      • JunctionDetectionLoss.training
    • RegularizationLoss
      • RegularizationLoss.forward()
      • RegularizationLoss.training
    • TotalLoss
      • TotalLoss.forward()
      • TotalLoss.forward_descriptors()
      • TotalLoss.training
    • TripletDescriptorLoss
      • TripletDescriptorLoss.descriptor_loss()
      • TripletDescriptorLoss.forward()
      • TripletDescriptorLoss.training
    • get_descriptor_loss_and_weight()
    • get_heatmap_loss_and_weight()
    • get_junction_loss_and_weight()
    • get_loss_and_weights()
    • heatmap_loss()
    • junction_detection_loss()
    • space_to_depth()
    • triplet_loss()
  • limap.line2d.SOLD2.model.lr_scheduler module
    • get_lr_scheduler()
  • limap.line2d.SOLD2.model.metrics module
    • AverageMeter
      • AverageMeter.average()
      • AverageMeter.update()
    • Metrics
      • Metrics.evaluate()
    • heatmap_precision
    • heatmap_recall
    • junction_pr
    • junction_precision
    • junction_recall
    • matching_score
    • nms_fast()
    • super_nms()
  • limap.line2d.SOLD2.model.model_util module
    • SOLD2Net
      • SOLD2Net.forward()
      • SOLD2Net.get_backbone()
      • SOLD2Net.get_descriptor_decoder()
      • SOLD2Net.get_heatmap_decoder()
      • SOLD2Net.get_junction_decoder()
      • SOLD2Net.training
    • get_model()
    • weight_init()
  • Module contents

Submodules

limap.line2d.SOLD2.experiment module

Main file to launch training and testing experiments.

limap.line2d.SOLD2.experiment.load_config(config_path)

Load configurations from a given yaml file.

limap.line2d.SOLD2.experiment.record_config(model_cfg, dataset_cfg, output_path)

Record dataset config to the log path.

limap.line2d.SOLD2.experiment.update_config(path, model_cfg=None, dataset_cfg=None)

Update configuration file from the resume path.

limap.line2d.SOLD2.sold2 module

class limap.line2d.SOLD2.sold2.SOLD2Detector(options=BaseDetectorOptions(set_gray=True, max_num_2d_segs=3000, do_merge_lines=False, visualize=False, weight_path=None))

Bases: BaseDetector

detect(camview)

Virtual method (for detector) - detect 2D line segments

Parameters:

view (limap.base.CameraView) – The limap.base.CameraView instance corresponding to the image

Returns:

line detections. Each row corresponds to x1, y1, x2, y2 and score.

Return type:

np.array of shape (N, 5)

detect_and_extract(camview)

Virtual method (for dual-functional class that can perform both detection and extraction) - Detect and extract on a single image

Parameters:

view (limap.base.CameraView) – The limap.base.CameraView instance corresponding to the image

Returns:

of shape (N, 5), line detections. Each row corresponds to x1, y1, x2, y2 and score. Computed from the detect method. descinfo: The features extracted from the function extract

Return type:

segs (np.array)

extract(camview, segs)

Virtual method (for extractor) - extract the features for the detected segments

Parameters:

• view (limap.base.CameraView) – The limap.base.CameraView instance corresponding to the image

• segs – np.array of shape (N, 5), line detections. Each row corresponds to x1, y1, x2, y2 and score. Computed from the detect method.

Returns:

The extracted feature

extract_heatmap(camview)

extract_heatmaps_all_images(folder, imagecols, skip_exists=False)

get_descinfo_fname(descinfo_folder, img_id)

Virtual method (for extractor) - Get the target filename of the extracted feature

Parameters:

• descinfo_folder (str) – The output folder

• img_id (int) – The image id

Returns:

target filename

Return type:

str

get_heatmap_fname(folder, img_id)

get_module_name()

Virtual method (need to be implemented) - return the name of the module

read_descinfo(descinfo_folder, img_id)

Virtual method (for extractor) - Read in the extracted feature. Dual function for save_descinfo.

Parameters:

• descinfo_folder (str) – The output folder

• img_id (int) – The image id

Returns:

The extracted feature

sample_descinfo_by_indexes(descinfo, indexes)

Virtual method (for dual-functional class that can perform both detection and extraction) - sample descriptors for a subset of images

Parameters:

• descinfo – The features extracted from the function extract.

• indexes (list[int]) – List of image ids for the subset.

save_descinfo(descinfo_folder, img_id, descinfo)

Virtual method (for extractor) - Save the extracted feature to the target folder

Parameters:

• descinfo_folder (str) – The output folder

• img_id (int) – The image id

• descinfo – The features extracted from the function extract

class limap.line2d.SOLD2.sold2.SOLD2Matcher(extractor, options=BaseMatcherOptions(topk=10, n_neighbors=20, n_jobs=1, weight_path=None))

Bases: BaseMatcher

get_module_name()

match_pair(descinfo1, descinfo2)

limap.line2d.SOLD2.sold2_wrapper module

class limap.line2d.SOLD2.sold2_wrapper.SOLD2LineDetector(device=None, cfg_path=None, weight_path=None)

Bases: object

compute_descinfo(segs, desc)

detect(input_image, saliency=False, scale_factor=None)

get_heatmap(input_image)

initialize_line_matcher()

match(img1, img2)

match_segs_with_descinfo(descinfo1, descinfo2)

match_segs_with_descinfo_topk(descinfo1, descinfo2, topk=10)

match_segs_with_descriptor(segs1, desc1, segs2, desc2)

segstosold2segs(segs)

sold2segstosegs(segs_sold2)

limap.line2d.SOLD2.train module

This file implements the training process and all the summaries

limap.line2d.SOLD2.train.convert_junc_predictions(predictions, grid_size, detect_thresh=0.015384615384615385, topk=300)

Convert torch predictions to numpy arrays for evaluation.

limap.line2d.SOLD2.train.customized_collate_fn(batch)

Customized collate_fn.

limap.line2d.SOLD2.train.plot_junction_detection(image_tensor, junc_pred_tensor, junc_pred_nms_tensor, junc_gt_tensor)

Plot the junction points on images.

limap.line2d.SOLD2.train.record_test_summaries(writer, epoch, scalars)

Record testing summaries.

limap.line2d.SOLD2.train.record_train_summaries(writer, global_step, scalars, images)

Record training summaries.

limap.line2d.SOLD2.train.restore_weights(model, state_dict, strict=True)

Restore weights in compatible mode.

limap.line2d.SOLD2.train.train_single_epoch(model, model_cfg, optimizer, loss_func, metric_func, train_loader, writer, epoch)

Train for one epoch.

limap.line2d.SOLD2.train.validate(model, model_cfg, loss_func, metric_func, val_loader, writer, epoch)

Validation.

Module contents