TY  - CONF
AU  - Carlo Gatta
AU  - Adriana Romero
AU  - Joost Van de Weijer
A2  - CVPRW
PY  - 2014//
TI  - Unrolling loopy top-down semantic feedback in convolutional deep networks
BT  - Workshop on Deep Vision: Deep Learning for Computer Vision
SP  - 498
EP  - 505
N2  - In this paper, we propose a novel way to perform top-down semantic feedback in convolutional deep networks for efficient and accurate image parsing. We also show how to add global appearance/semantic features, which have shown to improve image parsing performance in state-of-the-art methods, and was not present in previous convolutional approaches. The proposed method is characterised by an efficient training and a sufficiently fast testing. We use the well known SIFTflow dataset to numerically show the advantages provided by our contributions, and to compare with state-of-the-art image parsing convolutional based approaches.
L1  - http://refbase.cvc.uab.es/files/GRW2014.pdf
UR  - http://dx.doi.org/10.1109/CVPRW.2014.80
N1  - LAMP; MILAB; 601.160; 600.079
ID  - Carlo Gatta2014
ER  -