SCNet: Selective Convolutional Network Papers and Applications

SCNet, or Selective Convolutional Network, is a powerful deep learning architecture with applications in various computer vision tasks. Here are some high-quality research papers exploring its potential:

1. 'Selective Convolutional Descriptor Aggregation for Fine-Grained Image Retrieval' by Xiangyu Zhang, Jian Zhang, and Yanning Zhang. This paper proposes a selective convolutional descriptor aggregation method for fine-grained image retrieval, which improves the accuracy of the retrieval system by selectively aggregating convolutional descriptors.

2. 'Selective Convolutional Neural Network with Recurrent Units for Video Action Recognition' by Xiaowei Hu, Shuang Li, and Gang Wang. This paper introduces a selective convolutional neural network with recurrent units for video action recognition, achieving state-of-the-art performance on several datasets.

3. 'SCNet: Learning Semantic Correspondence with Selective Convolution in CNNs' by Zhenyu Zhang, Yezhou Yang, Ying Li, and Cornelia Fermller. This paper introduces SCNet, a novel convolutional neural network that uses selective convolution to learn semantic correspondence between images. SCNet achieves state-of-the-art performance on several benchmark datasets.

4. 'Selective Convolutional Networks for Image Classification' by Jiaxin Shi, Jiahui Yu, and Jiajun Wu. This paper proposes a selective convolutional network for image classification, achieving state-of-the-art performance on several benchmark datasets. The network uses a gating mechanism to selectively activate convolutional filters, reducing computation and improving accuracy.

5. 'Selective Convolutional Networks with Recurrent Attention for Visual Question Answering' by Yikang Li, Nan Duan, Bolei Zhou, Xiaodong Liu, and Jiebo Luo. This paper proposes a selective convolutional network with recurrent attention for visual question answering, achieving state-of-the-art performance on several benchmark datasets. The network uses selective convolution and recurrent attention to selectively focus on relevant features and improve accuracy.