Recurrent Convolutional Neural Networks for Scene Labeling

By Pedro H. O. Pinheiro, Ronan Collobert. 2014

731 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Fully convolutional networks for semantic segmentation (2014, methodology): “Patchwise training is common [27, 2, 7, 28, 9], but lacks the efficiency of fully convolutional training.”
The Cityscapes Dataset for Semantic Urban Scene Understanding (2016, background): “of pixel-wise semantic labels, in particular through CNNs [4,39,44,67] or Recurrent Neural Networks (RNNs) [8,52].”
Convolutional Pose Machines (2016, methodology): “In contrast, methods based on a sequential prediction framework [29] learn an implicit spatial model with potentially complex interactions between variables by directly training an inference procedure, as in [22, 25, 31, 41].”
Scene Parsing through ADE20K Dataset (2017, background): “With the success of convolutional neural networks (CNN) for image classification [16], there is growing interest for semantic image parsing using CNNs with dense output, such as the multiscale CNN [11], recurrent CNN [25], fully CNN [19], deconvolutional neural networks [24], encoder-decoder SegNet ”
Conditional Random Fields as Recurrent Neural Networks (2015, methodology): “in disadvantage of this strategy is that errors in the initial proposals (e.g: super-pixels) may lead to poor predictions, no matter how good the feature extraction process is. Pinheiro and Collobert [46] employed an RNN to model the spatial dependencies during scene parsing. In contrast to their ap”
Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation (2018, background): “, image pyramid) [18, 16, 58, 44, 11, 9] or those that adopt probabilistic graphical models (such as DenseCRF [37] with efficient inference algorithm [2]) [8, 4, 82, 44, 48, 55, 63, 34, 72, 6, 7, 9].”
Rethinking Atrous Convolution for Semantic Image Segmentation (2017, methodology): “First, the DCNN is applied to an image pyramid to extract features for each scale input [19, 17, 59, 46, 11, 10] where objects at different scales become prominent at different feature maps.”
Predicting Depth, Surface Normals and Semantic Labels with a Common Multi-scale Convolutional Architecture (2014, methodology): “[28] use a recurrent convolutional network in which each iteration incorporates progressively more context, by combining a more coarsely-sampled image input along with the local prediction from the previous iteration.”
How to Construct Deep Recurrent Neural Networks (2013, background): “One possible way to address this difﬁculty is to introduce shortcut connections (see, e.g., Raiko et al. , 2012) An approach similar to the deep hidden-to-hidden transition has been proposed recently by Pinheiro and Collobert (2014) in the context of parsing a static scene.”
Efficient Piecewise Training of Deep Structured Models for Semantic Segmentation (2015, neutral): “[38] 77.”
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation (2015, background): “Another approach using recurrent neural networks [39] merges several low resolution predictions to create input image resolution predictions.”
Stacked Hourglass Networks for Human Pose Estimation (2016, background): “Our work is closely connected to fully convolutional networks [23] and other designs that process spatial information in multiple scales for dense prediction [15,30,31,32,33,34,35,36,37,38].”
Feedforward semantic segmentation with zoom-out features (2014, background): “It is well known that training on imbalanced data without taking precautions can lead to poor results [10, 30, 22].”
Accurate De Novo Prediction of Protein Contact Map by Ultra-Deep Learning Model (2016, background): “ation of our deep learning model for contact prediction. Meanwhile, L is the sequence length of one protein under prediction. Figure 1 illustrates our deep neural network model for contact prediction [28]. Different from previous supervised learning approaches for contact prediction that employ only”