Scene labeling with LSTM recurrent neural networks

By Wonmin Byeon, Thomas Breuel, Federico Raue, Marcus Liwicki. 2015

365 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Model Architectures

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

In Model Architectures

Citations in the corpus

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].”
Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks (2015, methodology): “We achieve state-of-the-art results on both PASCAL VOC and COCO, and find our proposed architecture is particularly effective at improving detection of small objects.”
Rethinking Atrous Convolution for Semantic Image Segmentation (2017, methodology): “There are other methods based on LSTM [31] to aggregate global context [46, 6, 79].”
NTU RGB+D: A Large Scale Dataset for 3D Human Activity Analysis (2016, background): “Recurrent Neural Networks (RNNs) and Long Short-Term Memory Networks (LSTMs) [14] have been shown to be among the most successful deep learning models to encode and learn sequential data in various applications [35, 8, 2, 16].”
Dual Attention Network for Scene Segmentation (2018, methodology): “The method based on 2D LSTM networks [1] is proposed to capture complex spatial dependencies on labels.”
Structural-RNN: Deep Learning on Spatio-Temporal Graphs (2015, background): “Several previous works build multiple networks and wire them together in order to capture some complex structure (or interactions) in the problem with promising results on applications such as activity detection, scene labeling, image captioning, and object detection [12, 5, 9, 16, 49, 61].”
Learning Affinity via Spatial Propagation Networks (2017, background): “Our architecture is also related to the multi-dimensional RNN or LSTM [30, 3, 8].”
Image Segmentation Using Deep Learning: A Survey (2020, methodology): “[84] performed per-pixel segmentation and classification of images of natural scenes using 2D LSTM”
FuseNet: Incorporating Depth into Semantic Segmentation via Fusion-Based CNN Architecture (2016, background): “[5] have presented long short term memory (LSTM) recurrent neural networks for natural scene images taking into account the complex spatial dependencies of labels.”
Zoom Better to See Clearer: Human Part Segmentation with Auto Zoom Net (2015, background): “Some recent works have applied recurrent networks for computer vision tasks (Socher et al., 2012; Pinheiro & Collobert , 2013; Byeon et al., 2015).”
The ApolloScape Open Dataset for Autonomous Driving and Its Application (2018, background): “In our case, because the vehicle velocity is unknown, transition of camera poses is learned from the training sequences, and in our experiments we show that the motion predicted from RNN is better than using a Kalman filter with a constant speed assumption, yielding further improvement over the esti”
Deep Learning Advances in Computer Vision with 3D Data (2017, positive): “This indication was later conﬁrmed by other works as well (e.g., Byeon et al. [2015] and Yan et al. [2016] successfully used spatial RNNs for semantic segmentation, while Bell et al. [2015] exploited them for object detection).”
LatentGNN: Learning Efficient Non-local Relations for Visual Recognition (2019, background): “Early works tend to incorporate recurrent neural network (RNN) module to capture the long-range dependencies by constructing a directed acyclic graph model in convolutional feature space (Byeon et al., 2015; Shuai et al., 2018) or semantic label space (Zheng et al., 2015).”
ACFNet: Attentional Class Feature Network for Semantic Segmentation (2019, background): “Conditional random field (CRF) [6, 5, 7], Markov random field (MRF) [25] and Recurrent Neural Networks (RNNs) [4, 32] are also widely used to exploit the long-range dependencies.”
Scene Segmentation With Dual Relation-Aware Attention Network (2020, methodology): “Another type of method [15], [16] captures the long-range dependencies with recurrent neural networks, thus modeling relational context among pixels.”
Dense Recurrent Neural Networks for Scene Labeling (2018, background): “The work of [6] explores the two-dimensional long short term memory (LSTM) networks for scene parsing by taking into account the complex spatial dependencies of pixels in an image.”
Object Detection With Deep Learning: A Review (2018, background): “To this end, spatiotemporal tubelets [230], optical flow [199] and LSTM [107] should be considered to fundamentally model object associations”
A Siamese Long Short-Term Memory Architecture for Human Re-identification (2016, methodology): “RNN architectures have been previously used in [6,47,69] to model the spatial dependency and extract more discriminative features for image classiﬁcation and scene labeling.”