Neural Decision Forests for Semantic Image Labelling

By Samuel Rota Bulò, Peter Kontschieder. 2014

147 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

In Model Architectures

Citations in the corpus

Dynamic Neural Networks: A Survey (2021, background): “a) SDT [93], [94], [95] adopts neural units as its routing nodes (blue nodes in Fig.”
SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation (2015, neutral): “5 n/a Neural Decision Forests [60] n/a 56.”
Black-Box vs. White-Box: Understanding Their Advantages and Weaknesses From a Practical Point of View (2019, methodology): “Although models fusing rule and CNN have been used in some applied context like aspect level sentiment analysis [158] and semantic image labeling [153], the extracted”
Deep Neural Decision Trees (2018, methodology): “Bul & Kontschieder (2014) proposed Neural Decision Forests (NDF) as an ensemble of neural decision trees, where the split functions are realised by randomized multi-layer perceptrons.”
Generalizing Pooling Functions in Convolutional Neural Networks: Mixed, Gated, and Tree (2015, background): “In the supervised setting, [4] incorporates multilayer perceptrons within decision trees, but simply uses trained perceptrons as splitting nodes in a decision forest; not only does this result in training processes that are separate (and thus more difficult to train than an integrated training proce”
Bayesian SegNet: Model Uncertainty in Deep Convolutional Encoder-Decoder Architectures for Scene Understanding (2015, neutral): “5 n/a Neural Decision Forests [27] n/a 56.”
Generalizing Pooling Functions in CNNs: Mixed, Gated, and Tree (2018, background): “In the supervised setting, [24] incorporates multilayer perceptrons within decision trees, but simply uses trained perceptrons as splitting nodes in a decision forest; not only does this result in training processes that are separate (and thus more difficult to train than an integrated training proc”
SegNBDT: Visual Decision Rules for Segmentation (2020, methodology): “While some [11, 22, 7] have tried to fuse decision-tree and deep-learning methods, the resulting models sacrifice both accuracy and interpretability.”
Deep Neural Decision Forests (2015, methodology): “The approach in [28] introduced a way to integrate multi-layer perceptrons as split functions, however, representations were learned only locally at split node level and independently among split nodes.”
Deciding How to Decide: Dynamic Routing in Artificial Neural Networks (2017, methodology): “More recently, Kontschieder et al. (2015) performed joint optimization of ANN and decision tree parameters, and Bulo & Kontschieder (2014) used randomized multi-layer networks to compute decision tree split functions.”
Integrating parametric and non-parametric models for scene labeling (2015, methodology): “[1] learned a more compact random forest by substituting the random split function with a stronger Neural Network.”
Monocular Depth Estimation Using Neural Regression Forest (2016, methodology): “Thus, we learn (Θ,W) by maximizing the log-likelihood of the training data, as in [20, 11].”
Adaptive Neural Trees (2018, methodology): “Method Feature learning? Grown? Path Routers Soft DT [13] 7 7 3 Soft DT2 [14] 7 3 7 Conv DT [15] 7 3 7 NDT [16] 7 3 3 NDT2 [17] 3 3 7 NDF [18] 3 3 7 CNet [24] 3 3 7 ANT (ours) 3 3 3”
End-to-End Learning of Decision Trees and Forests (2019, background): “Rota Bulo and Kontschieder (2014) also apply a greedy strategy to learn neural networks for each split node and hence learn the structure of the tree.”
Deep Convolutional Neural Networks for Image Classification: A Comprehensive Review (2017, methodology): “Furthermore, inspired by Bulo and Kontschieder (2014), who incorporated MLPs with decision trees, Lee et al. (2016) used a binary decision tree to learn a combination of previously learned individual pooling filters.”
Deformable MR Prostate Segmentation via Deep Feature Learning and Sparse Patch Matching (2015, methodology): “To make the learned feature representation discriminative [33], [34], the supervised fine-tuning is often adopted by stacking another classification output layer on the top of the encoding part of the SSAE, as shown in red dashed box of Fig.”