Meta tags:

Headings (most frequently used words):

abstract, bibtex, acknowledgments,

Text of the page (most frequently used words):
the (13), and (8), adaptive (5), neighbourhood (5), consensus (5), with (4), cvpr (3), for (3), correspondence (3), networks (3), this (3), anc (3), net (3), shuda (2), kai (2), han (2), theo (2), costain (2), henry (2), howard (2), jenkins (2), victor (2), prisacariu (2), vision (2), 2020 (2), paper (2), task (2), dense (2), visual (2), intra (2), class (2), variations (2), annotations (2), propose (2), network (2), that (2), can (2), end (2), our (2), module (2), robust (2), matching (2), one (2), webpage, template, borrowed, from, split, brain, autoencoders, 2017, gratefully, acknowledge, support, european, commission, project, multiple, actors, virtual, empathiccaregiver, elder, movecare, epsrc, programme, grant, seebibyte, m013774, acknowledgments, inproceedings, li2020correspondence, author, title, booktitle, ieee, conference, computer, pattern, recognition, year, bibtex, tackle, establishing, correspondences, between, images, containing, objects, same, category, challenging, due, large, lack, pixel, level, convolutional, neural, architecture, called, trained, sparse, key, point, handle, challenge, core, proposed, non, isotropic, convolution, kernel, which, forms, building, block, also, introduce, simple, efficient, multi, scale, self, similarity, make, learned, feature, furthermore, novel, orthogonal, loss, enforce, constraint, thoroughly, evaluate, effectiveness, method, various, benchmarks, where, substantially, outperforms, state, art, methods, abstract, code, pytorch, indicates, equal, contribution, department, engineering, science, university, oxford, active, lab, geometry, group,

Text of the page (random words):
correspondence networks with adaptive neighbourhood consensus correspondence networks with adaptive neighbourhood consensus shuda li 1 kai han 2 theo w costain 1 henry howard jenkins 1 victor prisacariu 1 1 active vision lab 2 visual geometry group department of engineering science university of oxford indicates equal contribution paper cvpr 2020 code pytorch abstract in this paper we tackle the task of establishing dense visual correspondences between images containing objects of the same category this is a challenging task due to large intra class variations and a lack of dense pixel level annotations we propose a convolutional neural network architecture called adaptive neighbourhood consensus network anc net that can be trained end to end with sparse key point annotations to handle this challenge at the core of anc net is our proposed non isotropic 4d convolution kernel which forms the building block for the adaptive neighbourhood consensus module for robust matching we also introduce a simple and efficient multi scale self similarity module in anc net to make the learned feature robust to intra class variations furthermore we propose a novel orthogonal loss that can enforce the one to one matching constraint we thoroughly evaluate the effectiveness of our method on various benchmarks where it substantially outperforms state of the art methods bibtex inproceedings li2020correspondence author shuda li and kai han and theo w costain and henry howard jenkins and victor prisacariu title correspondence networks with adaptive neighbourhood consensus booktitle ieee conference on computer vision and pattern recognition cvpr year 2020 acknowledgments we gratefully acknowledge the support of the european commission project multiple actors virtual empathiccaregiver for the elder movecare and the epsrc programme grant seebibyte ep m013774 1 webpage template borrowed from split brain autoencoders cvpr 2017