Rotate to Attend: Convolutional Triplet Attention Module

Diganta Misra, Trikay Nalamada,
Ajay Uppili Arasanipalai, Qibin Hou

     

Attention in Computer Vision

Squeeze-Excite Networks

First to efficiently model cross-channel relationship in feature maps by learning channel-specific weights

CBAM

Fused channel and spatial attenion into a single module for visual recognition.

Global-Context Networks

Combines channel attention with non-local block to learn long-range dependancies

Shortcomings

  • Treat Dimensions Independently
  • Computationally Expensive
  • Information Bottlenecks

Cross-Dimensional Interaction

Triplet Attention

(a) Squeeze Excitation (b) CBAM; (c) Global Context Module; (d) Triplet Attention (ours)

Triplet Attention Structural Design

Results

Image Classification

Object Detection

GradCAM

Code and Pretrained Models:

https://landskapeai.github.io/publication/triplet/

References

  1. Yue Cao, Jiarui Xu, Stephen Lin, Fangyun Wei, and HanHu. Gcnet: Non-local networks meet squeeze-excitation networks and beyond. In Proceedings of the IEEE International Conference on Computer Vision Workshops, pages 0–0, 2019
  2. Jie Hu, Li Shen, and Gang Sun. Squeeze-and-excitation networks. 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Jun 2018
  3. Sanghyun Woo, Jongchan Park, Joon-Young Lee, and InSo Kweon. Cbam: Convolutional block attention module.In The European Conference on Computer Vision (ECCV),September 2018.
  4. Yunpeng Chen, Yannis Kalantidis, Jianshu Li, Shuicheng Yan, and Jiashi Feng. Aˆ2-nets: Double attention networks. In Advances in Neural Information Processing Systems, pages 352–361, 2018.
  5. Zilin Gao, Jiangtao Xie, Qilong Wang, and Peihua Li. Global second-order pooling convolutional networks. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pages 3024–3033, 2019