Label Assignment Matters: A Gaussian Assignment Strategy for Tiny Object Detection

doi:10.1109/TGRS.2024.3430071

Label Assignment Matters: A Gaussian Assignment Strategy for Tiny Object Detection

2024-01-01

4

PRE

AI

DOI：10.1109/TGRS.2024.3430071

Original

Original request for help

Save

Abstract

En 中文

Recently, impressive improvements have been achieved in general object detection. However, tiny object detection remains a very challenging problem since tiny objects only occupy a few pixels. Consequently, the label assignment strategies used in general object detectors are not suitable for tiny object detection, because these algorithms tend to assign few or even no positive samples for tiny objects. In this article, we propose a simple yet effective Gaussian assignment (GA) strategy to solve this problem. Specifically, we first model the bounding boxes as 2-D Gaussian distributions and then encode training samples with a threshold. This strategy can assign more high-quality positive samples for tiny objects and adjust the weight of positive samples to balance the contribution from different-size objects. Extensive experiments on four tiny object detection datasets show that the proposed strategy significantly and consistently improves the performance of single-stage tiny object detectors. In particular, with our strategy, we bridge the performance gap between single-stage and state-of-the-art multistage detectors on the AI-TOD dataset (24.2% versus 24.8% in mAP) while maintaining the inference speed. The code is available at https://github.com/zf020114/GaussianAssignment.

Keywords:

Detectors

Geoscience and remote sensing

Gaussian distribution

Annotations

Object detection

Heating systems

Feature extraction

Deep convolution neural networks

Gaussian distributions

tiny object detection

Label Assignment Matters: A Gaussian Assignment Strategy for Tiny Object Detection

Abstract

Journal

Organization