Pseudo-IoU: Improving Label Assignment in Anchor-Free Object Detection

TL;DR

This paper introduces Pseudo-IoU, a simple metric that enhances label assignment in anchor-free object detection, leading to improved accuracy without additional computational costs.

Contribution

The paper proposes Pseudo-IoU, a novel metric that standardizes label assignment in anchor-free detectors, bridging the performance gap with anchor-based methods.

Findings

Consistent performance improvements on PASCAL VOC and MSCOCO benchmarks.

Achieves state-of-the-art results among anchor-free detectors without extra complexity.

No additional computational cost or parameters introduced.

Abstract

Current anchor-free object detectors are quite simple and effective yet lack accurate label assignment methods, which limits their potential in competing with classic anchor-based models that are supported by well-designed assignment methods based on the Intersection-over-Union~(IoU) metric. In this paper, we present \textbf{Pseudo-Intersection-over-Union~(Pseudo-IoU)}: a simple metric that brings more standardized and accurate assignment rule into anchor-free object detection frameworks without any additional computational cost or extra parameters for training and testing, making it possible to further improve anchor-free object detection by utilizing training samples of good quality under effective assignment rules that have been previously applied in anchor-based methods. By incorporating Pseudo-IoU metric into an end-to-end single-stage anchor-free object detection framework, we observe consistent improvements in their performance on general object detection benchmarks such as PASCAL VOC and MSCOCO. Our method (single-model and single-scale) also achieves comparable performance to other recent state-of-the-art anchor-free methods without bells and whistles. Our code is based on mmdetection toolbox and will be made publicly available at https://github.com/SHI-Labs/Pseudo-IoU-for-Anchor-Free-Object-Detection.