Rotate to Attend: Convolutional Triplet Attention Module

TL;DR

This paper introduces a lightweight triplet attention module that captures cross-dimensional dependencies efficiently, improving performance in image classification and object detection tasks with minimal computational overhead.

Contribution

The paper proposes a novel triplet attention mechanism with a three-branch structure that effectively models inter-dimensional dependencies and can be easily integrated into existing networks.

Findings

Improves accuracy on ImageNet-1k classification

Enhances object detection performance on MSCOCO and PASCAL VOC

Maintains low computational cost and high efficiency

Abstract

Benefiting from the capability of building inter-dependencies among channels or spatial locations, attention mechanisms have been extensively studied and broadly used in a variety of computer vision tasks recently. In this paper, we investigate light-weight but effective attention mechanisms and present triplet attention, a novel method for computing attention weights by capturing cross-dimension interaction using a three-branch structure. For an input tensor, triplet attention builds inter-dimensional dependencies by the rotation operation followed by residual transformations and encodes inter-channel and spatial information with negligible computational overhead. Our method is simple as well as efficient and can be easily plugged into classic backbone networks as an add-on module. We demonstrate the effectiveness of our method on various challenging tasks including image classification on ImageNet-1k and object detection on MSCOCO and PASCAL VOC datasets. Furthermore, we provide extensive in-sight into the performance of triplet attention by visually inspecting the GradCAM and GradCAM++ results. The empirical evaluation of our method supports our intuition on the importance of capturing dependencies across dimensions when computing attention weights. Code for this paper can be publicly accessed at https://github.com/LandskapeAI/triplet-attention