SYNTOLOGY HomeExplorerAtlasCodeMethodologyAboutDevelopersFeedPricing
Paper · 1703.05593 · 2017

Convolutional neural network architecture for geometric matching

arXiv · PDF · Open in the Atlas

Code that ran

We lifted 14 functions out of this paper's own repositories and ran 6 of them in a sandbox. "Ran" means the function executed on a synthesized input and returned a value. It is not a reproduction of the paper's results.

RepositoryRoleRan
ignacio-rocco/cnngeometric_pytorch pwc_unofficial 5 of 12
Semanti1/cnngeometric_pytorch reimplementation 1 of 1
yoyongbo/test reimplementation 0 of 1
FunctionStatusWhere it lives
correct_keypoints Ran Semanti1/cnngeometric_pytorch/eval_pf.py
pointer only (licence: NONE) · get_code("cac8257f27511997")
default_collate Ran ignacio-rocco/cnngeometric_pytorch/util/dataloader.py
code served (permissive licence) · get_code("eb5ca9515b0fe778")
featureL2Norm Ran ignacio-rocco/cnngeometric_pytorch/model/cnn_geometric_model.py
code served (permissive licence) · get_code("cf9656ffec66503c")
normalize_axis Ran ignacio-rocco/cnngeometric_pytorch/geotnf/point_tnf.py
code served (permissive licence) · get_code("2fa12cb30a255583")
str_to_bool Ran ignacio-rocco/cnngeometric_pytorch/util/torch_util.py
code served (permissive licence) · get_code("8d87f08966a2f30c")
unnormalize_axis Ran ignacio-rocco/cnngeometric_pytorch/geotnf/point_tnf.py
code served (permissive licence) · get_code("f78be05f4ea43f88")
PointsToUnitCoords Not yet run ignacio-rocco/cnngeometric_pytorch/geotnf/point_tnf.py
code served (permissive licence) · get_code("17a7aebbcf3f2278")
correct_keypoints Not yet run yoyongbo/test/eval_pf.py
pointer only (licence: NONE) · get_code("db9d075e05b7e7b3")
expand_dim Not yet run ignacio-rocco/cnngeometric_pytorch/util/torch_util.py
code served (permissive licence) · get_code("82adf7750dd7a3e7")
homography_mat_from_4_pts Not yet run ignacio-rocco/cnngeometric_pytorch/geotnf/transformation.py
code served (permissive licence) · get_code("d5dfacfde93f35b2")
normalize_image Not yet run ignacio-rocco/cnngeometric_pytorch/image/normalization.py
code served (permissive licence) · get_code("6afbb61727662c3f")
pck Not yet run ignacio-rocco/cnngeometric_pytorch/util/eval_util.py
code served (permissive licence) · get_code("389ddc75e927381b")
pin_memory_batch Not yet run ignacio-rocco/cnngeometric_pytorch/util/dataloader.py
code served (permissive licence) · get_code("6d8c67dc116f1f8a")
read_flo_file Not yet run ignacio-rocco/cnngeometric_pytorch/geotnf/flow.py
code served (permissive licence) · get_code("502159ccb250f467")

Repositories linked to this paper

Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.

Abstract

We address the problem of determining correspondences between two images in agreement with a geometric model such as an affine or thin-plate spline transformation, and estimating its parameters. The contributions of this work are three-fold. First, we propose a convolutional neural network architecture for geometric matching. The architecture is based on three main components that mimic the standard steps of feature extraction, matching and simultaneous inlier detection and model parameter estimation, while being trainable end-to-end. Second, we demonstrate that the network parameters can be trained from synthetically generated imagery without the need for manual annotation and that our matching layer significantly increases generalization capabilities to never seen before images. Finally, we show that the same model can perform both instance-level and category-level matching giving state-of-the-art results on the challenging Proposal Flow dataset.

For agents

The same record, over MCP at https://syntology.ai/mcp:

get_harvested_code_for_paper("1703.05593")
get_code_for_paper("1703.05593")
have("1703.05593")

Connect an agent — have() is free.