Learning fine-grained features via a CNN Tree for Large-scale Classification

By Zhenhua Wang, Xingxing Wang, G. Wang. 2015

52 citations recorded by the source corpus.

We propose a novel approach to enhance the discriminability of Convolutional Neural Networks (CNN). The key idea is to build a tree structure that could progressively learn fine-grained features to distinguish a subset of classes, by learning features only among these classes. Such features are expected to be more discriminative, compared to features learned for all the classes. We develop a new algorithm to effectively learn the tree structure from a large number of classes. Experiments on large-scale image classification tasks demonstrate that our method could boost the performance of a given basic CNN model. Our method is quite general, hence it can potentially be used in combination with many other deep learning models.

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

1. Jianfei Cai
2. G. Wang
3. Jiashi Feng
4. Shuicheng Yan
5. Ming-Hsuan Yang
6. Luc Van Gool
7. Jiuxiang Gu
8. Yi Yang

Citations in the corpus

• Recent advances in convolutional neural networks (2015, background): “Similarly, [95] builds a tree structure to learn fine-grained features for subcategory recognition.”
• A survey on deep learning-based fine-grained object classification and semantic segmentation (2017, methodology): “This figure is from the original paper [32].”
• Diversified Visual Attention Networks for Fine-Grained Object Classification (2016, background): “Motivated by the observation that a class is usually confused with a few other classes in multi-class classification, a CNN Tree [28] was built to progressively learn find-grained features to distinguish a subset of classes by learning features only among these classes.”