computer-vision-1

Dense Pixel Correspondence Estimation

3260 papers • 126 benchmarks • 313 datasets

This task has no description! Would you like to contribute one?

(Image credit: Papersgraph)

Benchmarks

These leaderboards are used to track progress in dense-pixel-correspondence-estimation-1

Trend

Dataset

Best Model

Actions

HPatches

KITTI 2012

KITTI 2015

Libraries

i

Use these libraries to find dense-pixel-correspondence-estimation-1 models and implementations

open-mmlab/mmflow

2 papers 887

Datasets

KITTI

HPatches

ETH3D

Subtasks

No subtasks available.

Most implemented papers

FlowNet 2.0: Evolution of Optical Flow Estimation with Deep Networks

T. Brox, Eddy Ilg, N. Mayer, Tonmoy Saikia, Margret Keuper, Alexey Dosovitskiy•Mon Dec 05 2016

The concept of end-to-end learning of optical flow is advanced and it work really well, and faster variants that allow optical flow computation at up to 140fps with accuracy matching the original FlowNet are presented.

3268

Content

ETH3D

PruneTruong/DenseMatching

2 papers 614

0

Paper Graph

PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume

Ming-Yu Liu, Jan Kautz, Xiaodong Yang, Deqing Sun•Wed Sep 06 2017

PWC-Net has been designed according to simple and well-established principles: pyramidal processing, warping, and the use of a cost volume, and outperforms all published optical flow methods on the MPI Sintel final pass and KITTI 2015 benchmarks.

2711 0

Paper Graph

DGC-Net: Dense Geometric Correspondence Network

M. Pollefeys, Esa Rahtu, Juho Kannala, Torsten Sattler, Iaroslav Melekhov, Aleksei Tiulpin•Thu Oct 18 2018

This paper proposes a coarse-to-fine CNN-based framework that can leverage the advantages of optical flow approaches and extend them to the case of large transformations providing dense and subpixel accurate estimates and proves that the model outperforms existing dense approaches.

130 0

Paper Graph

Optical Flow Estimation Using a Spatial Pyramid Network

Michael J. Black, Anurag Ranjan•Wed Nov 02 2016

The Spatial Pyramid Network (SPyNet) is much simpler and 96% smaller than FlowNet in terms of model parameters, which makes it more efficient and appropriate for embedded applications.

1342 0

Paper Graph

Learning Accurate Dense Correspondences and When to Trust Them

L. Gool, Martin Danelljan, Radu Timofte, Prune Truong•Mon Jan 04 2021

This work aims to estimate a dense flow field relating two images, coupled with a robust pixel-wise confidence map indicating the reliability and accuracy of the prediction, and develops a flexible probabilistic approach that jointly learns the flow prediction and its uncertainty.

155 0

Paper Graph

GLU-Net: Global-Local Universal Network for Dense Flow and Correspondences

Martin Danelljan, Radu Timofte, Prune Truong•Tue Dec 10 2019

This work proposes a universal network architecture that is directly applicable to all the aforementioned dense correspondence problems, and achieves both high accuracy and robustness to large displacements by investigating the combined use of global and local correlation layers.

228 0

Paper Graph

DeepMatching: Hierarchical Deformable Dense Matching

Philippe Weinzaepfel, Jérôme Revaud, Zaïd Harchaoui, C. Schmid•Wed Jun 24 2015

A novel matching algorithm, called DeepMatching, to compute dense correspondences between images, which outperforms the state-of-the-art algorithms and shows excellent results in particular for repetitive textures.

282 0

Paper Graph

Space-Time Correspondence as a Contrastive Random Walk

Alexei A. Efros, A. Jabri, Andrew Owens•Wed Jun 24 2020

Despite its simplicity, the method outperforms the self-supervised state-of-the-art on a variety of label propagation tasks involving objects, semantic parts, and pose.

336 0

Paper Graph

GOCor: Bringing Globally Optimized Correspondence Volumes into Your Neural Network

L. Gool, Martin Danelljan, Radu Timofte, Prune Truong•Tue Sep 15 2020

The proposed GOCor module is a fully differentiable dense matching module, acting as a direct replacement to the feature correlation layer, capable of effectively learning spatial matching priors to resolve further matching ambiguities.

84 0

Paper Graph

COTR: Correspondence Transformer for Matching Across Images

Eduard Trulls, A. Tagliasacchi, K. M. Yi, Wei Jiang, J. Hosang•Wed Mar 24 2021

A novel framework for finding correspondences in images based on a deep neural network that, given two images and a query point in one of them, finds its correspondence in the other, yielding a multiscale pipeline able to provide highly-accurate correspondences.

303 0

Paper Graph

Adding a benchmark result helps the community track progress.