Mutual CRF-GNN for Few-shot Learning

By Shixiang Tang, Dapeng Chen, Lei Bai, Kaijian Liu, Yixiao Ge, Wanli Ouyang, Hong Kong. 2021

60 citations recorded by the source corpus.

Graph-neural-networks (GNN) is a rising trend for fewshot learning. A critical component in GNN is the affinity. Typically, affinity in GNN is mainly computed in the feature space, e.g., pairwise features, and does not take fully advantage of semantic labels associated to these features. In this paper, we propose a novel Mutual CRF-GNN (MCGN). In this MCGN, the labels and features of support data are used by the CRF for inferring GNN affinities in a principled and probabilistic way. Specifically, we construct a Conditional Random Field (CRF) conditioned on labels and features of support data to infer a affinity in the label space. Such affinity is fed to the GNN as the node-wise affinity. GNN and CRF mutually contributes to each other in MCGN. For GNN, CRF provides valuable affinity information. For CRF, GNN provides better features for inferring affinity. Experimental results show that our approach outperforms stateof-the-arts on datasets miniImageNet, tieredImageNet, and CIFAR-FS on both 5-way 1-shot and 5-way 5-shot settings.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

1. Jinguo Zhu
2. Shijie Yu
3. Dapeng Chen
4. Shixiang Tang
5. Wanli Ouyang
6. Bo Li
7. Boyu Chen
8. Peixia Li

In Computer Vision and Pattern Recognition

1. Wanli Ouyang
2. Xiaogang Wang
3. Dacheng Tao
4. Luc Van Gool
5. Bo Li
6. Wei Wu
7. Dapeng Chen
8. Yi Yang

Citations in the corpus

• Layerwise Optimization by Gradient Decomposition for Continual Learning (2021, positive): “Layerwise gradient update strategy is also applied in meta-learning for rapid loss convergence on transfer learning [13, 14, 48] and few-shot learning [49, 28].”
• Backbone is All Your Need: A Simplified Architecture for Visual Object Tracking (2022, background): “The transformer backbone is a one-branch backbone instead of a Siamese network, consistent with the backbones used in many vision tasks, e.g., image classification [21,15,40,49], object detection [38], semantic segmentation [20,53], depth estimation [25,43], etc .”
• Complementary Relation Contrastive Distillation (2021, background): “It has a wide range of applications in the industry especially when a neural network needs to be efficiently deployed on devices with limited computational resources [9, 54, 38].”