Tampere University of Technology

TUTCRIS Research Portal

Graph-boosted attentive network for semantic body parsing

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Standard

Graph-boosted attentive network for semantic body parsing. / Wang, Tinghuai; Wang, Huiling.

Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing - 28th International Conference on Artificial Neural Networks, 2019, Proceedings. ed. / Igor V. Tetko; Pavel Karpov; Fabian Theis; Vera Kurková. Springer Verlag, 2019. p. 267-280 (Lecture Notes in Computer Science; Vol. 11729).

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Harvard

Wang, T & Wang, H 2019, Graph-boosted attentive network for semantic body parsing. in IV Tetko, P Karpov, F Theis & V Kurková (eds), Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing - 28th International Conference on Artificial Neural Networks, 2019, Proceedings. Lecture Notes in Computer Science, vol. 11729, Springer Verlag, pp. 267-280, International Conference on Artificial Neural Networks, Munich, Germany, 17/09/19. https://doi.org/10.1007/978-3-030-30508-6_22

APA

Wang, T., & Wang, H. (2019). Graph-boosted attentive network for semantic body parsing. In I. V. Tetko, P. Karpov, F. Theis, & V. Kurková (Eds.), Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing - 28th International Conference on Artificial Neural Networks, 2019, Proceedings (pp. 267-280). (Lecture Notes in Computer Science; Vol. 11729). Springer Verlag. https://doi.org/10.1007/978-3-030-30508-6_22

Vancouver

Wang T, Wang H. Graph-boosted attentive network for semantic body parsing. In Tetko IV, Karpov P, Theis F, Kurková V, editors, Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing - 28th International Conference on Artificial Neural Networks, 2019, Proceedings. Springer Verlag. 2019. p. 267-280. (Lecture Notes in Computer Science). https://doi.org/10.1007/978-3-030-30508-6_22

Author

Wang, Tinghuai ; Wang, Huiling. / Graph-boosted attentive network for semantic body parsing. Artificial Neural Networks and Machine Learning – ICANN 2019: Image Processing - 28th International Conference on Artificial Neural Networks, 2019, Proceedings. editor / Igor V. Tetko ; Pavel Karpov ; Fabian Theis ; Vera Kurková. Springer Verlag, 2019. pp. 267-280 (Lecture Notes in Computer Science).

Bibtex - Download

@inproceedings{c841b560f9bb482db5b65966bbc7e3f9,
title = "Graph-boosted attentive network for semantic body parsing",
abstract = "Human body parsing remains a challenging problem in natural scenes due to multi-instance and inter-part semantic confusions as well as occlusions. This paper proposes a novel approach to decomposing multiple human bodies into semantic part regions in unconstrained environments. Specifically we propose a convolutional neural network (CNN) architecture which comprises of novel semantic and contour attention mechanisms across feature hierarchy to resolve the semantic ambiguities and boundary localization issues related to semantic body parsing. We further propose to encode estimated pose as higher-level contextual information which is combined with local semantic cues in a novel graphical model in a principled manner. In this proposed model, the lower-level semantic cues can be recursively updated by propagating higher-level contextual information from estimated pose and vice versa across the graph, so as to alleviate erroneous pose information and pixel level predictions. We further propose an optimization technique to efficiently derive the solutions. Our proposed method achieves the state-of-art results on the challenging Pascal Person-Part dataset.",
author = "Tinghuai Wang and Huiling Wang",
note = "jufoid=62555",
year = "2019",
doi = "10.1007/978-3-030-30508-6_22",
language = "English",
isbn = "9783030305079",
series = "Lecture Notes in Computer Science",
publisher = "Springer Verlag",
pages = "267--280",
editor = "Tetko, {Igor V.} and Pavel Karpov and Fabian Theis and Vera Kurkov{\'a}",
booktitle = "Artificial Neural Networks and Machine Learning – ICANN 2019",
address = "Germany",

}

RIS (suitable for import to EndNote) - Download

TY - GEN

T1 - Graph-boosted attentive network for semantic body parsing

AU - Wang, Tinghuai

AU - Wang, Huiling

N1 - jufoid=62555

PY - 2019

Y1 - 2019

N2 - Human body parsing remains a challenging problem in natural scenes due to multi-instance and inter-part semantic confusions as well as occlusions. This paper proposes a novel approach to decomposing multiple human bodies into semantic part regions in unconstrained environments. Specifically we propose a convolutional neural network (CNN) architecture which comprises of novel semantic and contour attention mechanisms across feature hierarchy to resolve the semantic ambiguities and boundary localization issues related to semantic body parsing. We further propose to encode estimated pose as higher-level contextual information which is combined with local semantic cues in a novel graphical model in a principled manner. In this proposed model, the lower-level semantic cues can be recursively updated by propagating higher-level contextual information from estimated pose and vice versa across the graph, so as to alleviate erroneous pose information and pixel level predictions. We further propose an optimization technique to efficiently derive the solutions. Our proposed method achieves the state-of-art results on the challenging Pascal Person-Part dataset.

AB - Human body parsing remains a challenging problem in natural scenes due to multi-instance and inter-part semantic confusions as well as occlusions. This paper proposes a novel approach to decomposing multiple human bodies into semantic part regions in unconstrained environments. Specifically we propose a convolutional neural network (CNN) architecture which comprises of novel semantic and contour attention mechanisms across feature hierarchy to resolve the semantic ambiguities and boundary localization issues related to semantic body parsing. We further propose to encode estimated pose as higher-level contextual information which is combined with local semantic cues in a novel graphical model in a principled manner. In this proposed model, the lower-level semantic cues can be recursively updated by propagating higher-level contextual information from estimated pose and vice versa across the graph, so as to alleviate erroneous pose information and pixel level predictions. We further propose an optimization technique to efficiently derive the solutions. Our proposed method achieves the state-of-art results on the challenging Pascal Person-Part dataset.

U2 - 10.1007/978-3-030-30508-6_22

DO - 10.1007/978-3-030-30508-6_22

M3 - Conference contribution

SN - 9783030305079

T3 - Lecture Notes in Computer Science

SP - 267

EP - 280

BT - Artificial Neural Networks and Machine Learning – ICANN 2019

A2 - Tetko, Igor V.

A2 - Karpov, Pavel

A2 - Theis, Fabian

A2 - Kurková, Vera

PB - Springer Verlag

ER -