Author name cluster

Gang Sun

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

1 author row

EAAI Journal 2023 Journal Article

RFIA-Net: Rich CNN-transformer network based on asymmetric fusion feature aggregation to classify stage I multimodality oesophageal cancer images

Zhicheng Zhou
Gang Sun
Long Yu
Shengwei Tian
Guangli Xiao
Junwen Wang
Shaofeng Zhou

Endoscopic images of oesophageal cancer have the characteristics of rich colours; furthermore, the small lesions are similar to the oesophageal wall tissue, and the pathological images have the characteristics of various staining methods, different shapes, and rich texture details. Aiming at the above characteristics and combining the unique advantages of convolutional architectures and the development of vision transformers in computer vision tasks, in this paper, for the stage I multimodality oesophageal cancer image classification task, we design an efficient hybrid architecture that leverages the local modelling capabilities and powerful semantic feature extraction capabilities of convolutional neural networks and the ability of transformers to extract global information. And combined with the structural reparameterization strategy to further improve the model expression. Specifically, our architecture consists of a feature extraction module and a feature enhancement module. In the feature enhancement module, we supplement the semantic information of each branch by continuously exchanging information between the two branches, which further improves the performance of the network. Furthermore, we propose an asymmetric fusion module that allows features to further enhance the feature relationships between different branches through spatial translation and channel swapping. Compared with networks such as ResNet-18, our proposed method achieves the best results for oesophageal cancer image classification on both tasks on the XJMU-XJU stage I multimodal oesophageal cancer dataset. The proposed method achieved an AUC of 0. 9973 and an ACC of 0. 9902 on the staging task and achieved a recall of 0. 9742 and an ACC of 0. 9750 on the differentiation task.

Details DOI

NeurIPS Conference 2018 Conference Paper

Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks

Jie Hu
Li Shen
Samuel Albanie
Gang Sun
Andrea Vedaldi

While the use of bottom-up local operators in convolutional neural networks (CNNs) matches well some of the statistics of natural images, it may also prevent such models from capturing contextual long-range feature interactions. In this work, we propose a simple, lightweight approach for better context exploitation in CNNs. We do so by introducing a pair of operators: gather, which efficiently aggregates feature responses from a large spatial extent, and excite, which redistributes the pooled information to local features. The operators are cheap, both in terms of number of added parameters and computational complexity, and can be integrated directly in existing architectures to improve their performance. Experiments on several datasets show that gather-excite can bring benefits comparable to increasing the depth of a CNN at a fraction of the cost. For example, we find ResNet-50 with gather-excite operators is able to outperform its 101-layer counterpart on ImageNet with no additional learnable parameters. We also propose a parametric gather-excite operator pair which yields further performance gains, relate it to the recently-introduced Squeeze-and-Excitation Networks, and analyse the effects of these changes to the CNN feature activation statistics.

PDF Details

IJCAI Conference 2015 Conference Paper

Adaptive Sharing for Image Classification

Li Shen
Gang Sun
Zhouchen Lin
Qingming Huang
Enhua Wu

In this paper, we formulate the image classification problem in a multi-task learning framework. We propose a novel method to adaptively share information among tasks (classes). Different from imposing strong assumptions or discovering specific structures, the key insight in our method is to selectively extract and exploit the shared information among classes while capturing respective disparities simultaneously. It is achieved by estimating a composite of two sets of parameters with different regularization. Besides applying it for learning classifiers on pre-computed features, we also integrate the adaptive sharing with deep neural networks, whose discriminative power can be augmented by encoding class relationship. We further develop two strategies for solving the optimization problems in the two scenarios. Empirical results demonstrate that our method can significantly improve the classification performance by transferring knowledge appropriately.

PDF Details