Recent works have shown that exploiting multi-scale representations deeply learned via convolutional neural networks (CNN) is of tremendous importance for accurate contour detection. This paper presents a novel approach for predicting contours which advances the state of the art in two fundamental aspects, i.e. multi-scale feature generation and fusion. Different from previous works directly considering multi-scale feature maps obtained from the inner layers of a primary CNN architecture, we introduce a hierarchical deep model which produces more rich and complementary representations. Furthermore, to refine and robustly fuse the representations learned at different scales, the novel Attention-Gated Conditional Random Fields (AG-CRFs) are proposed. The experiments ran on two publicly available datasets (BSDS500 and NYUDv2) demonstrate the effectiveness of the latent AG-CRF model and of the overall hierarchical framework.

Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction / Xu, Dan; Ouyang, Wanli; Alameda Pineda, Xavier; Ricci, Elisa; Wang, Xiaogang; Sebe, Niculae. - (2017), pp. 3964-3973. (Intervento presentato al convegno NIPS tenutosi a Long Beach, CA, USA nel December 2017).

Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction

Dan Xu;Xavier Alameda-Pineda;Elisa Ricci;Nicu Sebe
2017-01-01

Abstract

Recent works have shown that exploiting multi-scale representations deeply learned via convolutional neural networks (CNN) is of tremendous importance for accurate contour detection. This paper presents a novel approach for predicting contours which advances the state of the art in two fundamental aspects, i.e. multi-scale feature generation and fusion. Different from previous works directly considering multi-scale feature maps obtained from the inner layers of a primary CNN architecture, we introduce a hierarchical deep model which produces more rich and complementary representations. Furthermore, to refine and robustly fuse the representations learned at different scales, the novel Attention-Gated Conditional Random Fields (AG-CRFs) are proposed. The experiments ran on two publicly available datasets (BSDS500 and NYUDv2) demonstrate the effectiveness of the latent AG-CRF model and of the overall hierarchical framework.
2017
Advances in Neural Information Processing Systems 30
Dutchess County, New York
Curran Associates, Inc.
Xu, Dan; Ouyang, Wanli; Alameda Pineda, Xavier; Ricci, Elisa; Wang, Xiaogang; Sebe, Niculae
Learning Deep Structured Multi-Scale Features using Attention-Gated CRFs for Contour Prediction / Xu, Dan; Ouyang, Wanli; Alameda Pineda, Xavier; Ricci, Elisa; Wang, Xiaogang; Sebe, Niculae. - (2017), pp. 3964-3973. (Intervento presentato al convegno NIPS tenutosi a Long Beach, CA, USA nel December 2017).
File in questo prodotto:
File Dimensione Formato  
6985-learning-deep-structured-multi-scale-features-using-attention-gated-crfs-for-contour-prediction.pdf

accesso aperto

Tipologia: Versione editoriale (Publisher’s layout)
Licenza: Tutti i diritti riservati (All rights reserved)
Dimensione 2.43 MB
Formato Adobe PDF
2.43 MB Adobe PDF Visualizza/Apri

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/11572/194357
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 79
  • ???jsp.display-item.citation.isi??? ND
social impact