Decoupling Category-wise Independence and Relevance with Self-attention for Multi-label Image Classification

By Luchen Liu, Sheng Guo, Weilin Huang, Matthew R. Scott. 2019

13 citations recorded by the source corpus.

Multi-label image classification has achieved remarkable progress thanks to deep convolutional neural networks (CNNs). In this paper, we propose a Decouple Network (DecoupleNet) which is an end-to-end CNN-based framework able to trade off class-level feature independence and relevance during training. The proposed DecoupleNet is able to decouple category-wise independence and relevance with image-level supervision. We design a category-wise space-to-depth module with a spatial pooling strategy to exploit more meaningful convolutional features. They are integrated with class-wise correlated information which is automatically learned via a new self-attention mechanism. We conduct extensive experiments on two large-scale benchmarks: the MS-COCO and the NUS-WIDE, where the proposed DecoupleNet obtains impressive performance compared favorably against the state-of-the-art methods on multi-label image classification.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Molecular Biology

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

1. Michal Drozdzal
2. Adriana Romero
3. Luis Villaseñor Pineda
4. Amaia Salvador
5. Yoshua Bengio
6. Dacheng Tao
7. Jure Leskovec

In Computer Vision and Pattern Recognition

1. Michal Drozdzal
2. Luis Villaseñor Pineda
3. Adriana Romero
4. Amaia Salvador
5. Jian Zhan
6. Su He
7. Guo-Liang Zhang
8. M.L. Luo

In Molecular Biology

1. Adriana Romero
2. Luis Villaseñor Pineda
3. Amaia Salvador
4. Michal Drozdzal

Citations in the corpus

• Elucidating image-to-set prediction: An analysis of models, losses and datasets (2019, methodology): “2019 [33] ResNet-101 n/a 82787 n/a 40504 77.”
• Efficient Multi-Scale Attention Module with Cross-Spatial Learning (2023, methodology): “It uses the matrix dotproduct operations aiming at capturing pixel-level pairwise relationship and highlighting global context for all pixels [25], [26].”