Trident Pyramid Networks: The importance of processing at the feature pyramid level for better object detection

TL;DR

This paper introduces Trident Pyramid Networks (TPN), a new neck architecture for feature pyramids in object detection, which improves accuracy by enabling deeper processing and better resource allocation at the pyramid level.

Contribution

The paper proposes a novel TPN neck architecture that allows for deeper processing and better balance between communication and self-processing, leading to improved detection performance.

Findings

TPN outperforms BiFPN by 0.5 AP on COCO.

Adding computation to TPN yields better results than increasing backbone complexity.

Tuning computation at the feature pyramid level is more effective for detection accuracy.

Abstract

Feature pyramids have become ubiquitous in multi-scale computer vision tasks such as object detection. Given their importance, a computer vision network can be divided into three parts: a backbone (generating a feature pyramid), a neck (refining the feature pyramid) and a head (generating the final output). Many existing networks operating on feature pyramids, named necks, are shallow and mostly focus on communication-based processing in the form of top-down and bottom-up operations. We present a new neck architecture called Trident Pyramid Network (TPN), that allows for a deeper design and for a better balance between communication-based processing and self-processing. We show consistent improvements when using our TPN neck on the COCO object detection benchmark, outperforming the popular BiFPN baseline by 0.5 AP, both when using the ResNet-50 and the ResNeXt-101-DCN backbone. Additionally, we empirically show that it is more beneficial to put additional computation into the TPN neck, rather than into the backbone, by outperforming a ResNet-101+FPN baseline with our ResNet-50+TPN network by 1.7 AP, while operating under similar computation budgets. This emphasizes the importance of performing computation at the feature pyramid level in modern-day object detection systems. Code is available at https://github.com/CedricPicron/TPN .