A CNN cascade for landmark guided semantic part segmentation

This paper proposes a CNN cascade for semantic part segmentation guided by pose-specifc information encoded in terms of a set of landmarks (or keypoints). There is large amount of prior work on each of these tasks separately, yet, to the best of our knowledge, this is the first time in literature th...

Full description

Bibliographic Details
Main Authors: Jackson, Aaron S., Valstar, Michel, Tzimiropoulos, Georgios
Format: Conference or Workshop Item
Published: 2016
Subjects:
Online Access:https://eprints.nottingham.ac.uk/37234/
_version_ 1848795418935164928
author Jackson, Aaron S.
Valstar, Michel
Tzimiropoulos, Georgios
author_facet Jackson, Aaron S.
Valstar, Michel
Tzimiropoulos, Georgios
author_sort Jackson, Aaron S.
building Nottingham Research Data Repository
collection Online Access
description This paper proposes a CNN cascade for semantic part segmentation guided by pose-specifc information encoded in terms of a set of landmarks (or keypoints). There is large amount of prior work on each of these tasks separately, yet, to the best of our knowledge, this is the first time in literature that the interplay between pose estimation and semantic part segmentation is investigated. To address this limitation of prior work, in this paper, we propose a CNN cascade of tasks that firstly performs landmark localisation and then uses this information as input for guiding semantic part segmentation. We applied our architecture to the problem of facial part segmentation and report large performance improvement over the standard unguided network on the most challenging face datasets. Testing code and models will be published online at http://cs.nott.ac.uk/~psxasj/.
first_indexed 2025-11-14T19:31:47Z
format Conference or Workshop Item
id nottingham-37234
institution University of Nottingham Malaysia Campus
institution_category Local University
last_indexed 2025-11-14T19:31:47Z
publishDate 2016
recordtype eprints
repository_type Digital Repository
spelling nottingham-372342020-05-04T18:17:41Z https://eprints.nottingham.ac.uk/37234/ A CNN cascade for landmark guided semantic part segmentation Jackson, Aaron S. Valstar, Michel Tzimiropoulos, Georgios This paper proposes a CNN cascade for semantic part segmentation guided by pose-specifc information encoded in terms of a set of landmarks (or keypoints). There is large amount of prior work on each of these tasks separately, yet, to the best of our knowledge, this is the first time in literature that the interplay between pose estimation and semantic part segmentation is investigated. To address this limitation of prior work, in this paper, we propose a CNN cascade of tasks that firstly performs landmark localisation and then uses this information as input for guiding semantic part segmentation. We applied our architecture to the problem of facial part segmentation and report large performance improvement over the standard unguided network on the most challenging face datasets. Testing code and models will be published online at http://cs.nott.ac.uk/~psxasj/. 2016-10-07 Conference or Workshop Item PeerReviewed Jackson, Aaron S., Valstar, Michel and Tzimiropoulos, Georgios (2016) A CNN cascade for landmark guided semantic part segmentation. In: ECCV 2016 Workshop, Geometry meets Deep Learning, 9 October 2016, Amsterdam, Netherlands. pose estimation landmark localisation semantic part seg- mentation faces
spellingShingle pose estimation
landmark localisation
semantic part seg- mentation
faces
Jackson, Aaron S.
Valstar, Michel
Tzimiropoulos, Georgios
A CNN cascade for landmark guided semantic part segmentation
title A CNN cascade for landmark guided semantic part segmentation
title_full A CNN cascade for landmark guided semantic part segmentation
title_fullStr A CNN cascade for landmark guided semantic part segmentation
title_full_unstemmed A CNN cascade for landmark guided semantic part segmentation
title_short A CNN cascade for landmark guided semantic part segmentation
title_sort cnn cascade for landmark guided semantic part segmentation
topic pose estimation
landmark localisation
semantic part seg- mentation
faces
url https://eprints.nottingham.ac.uk/37234/