Res2Net: A New Multi-Scale Backbone Architecture
Nankai University · University of California, Merced · +1 more institution
Abstract
Representing features at multiple scales is of great importance for numerous vision tasks. Recent advances in backbone convolutional neural networks (CNNs) continually demonstrate stronger multi-scale representation ability, leading to consistent performance gains on a wide range of applications. However, most existing methods represent the multi-scale features in a layer-wise manner. In this paper, we propose a novel building block for CNNs, namely Res2Net, by constructing hierarchical residual-like connections within one single residual block. The Res2Net represents multi-scale features at a granular level and increases the range of receptive fields for each network layer. The proposed Res2Net block can be…
Citation impact
- FWCI
- 116.66
- Percentile
- 100%
- References
- 115
Authors
6Topics & keywords
- Computer science
- Block (permutation group theory)
- Artificial intelligence
- Residual
- Backbone network
- Convolutional neural network
- Object detection
- Pattern recognition (psychology)
- Sustainable cities and communities