Small Bird Detection using YOLOv7 with Test-Time Augmentation

TL;DR

This paper enhances small bird detection by applying YOLOv7 with test-time augmentation techniques, significantly improving accuracy in a competitive challenge.

Contribution

It introduces a combination of input resolution increase, multiscale inference, flipped images, and weighted boxes fusion to improve small object detection performance.

Findings

Achieved a public AP of 0.732 and private AP of 27.2 at IoU=0.5.

Significant improvements in detection accuracy demonstrated.

Method outperformed other entries in the challenge.

Abstract

In this paper, we propose a method specifically aimed at improving small bird detection for the Small Object Detection Challenge for Spotting Birds 2023. Utilizing YOLOv7 model with test-time augmentation, our approach involves increasing the input resolution, incorporating multiscale inference, considering flipped images during the inference process, and employing weighted boxes fusion to merge detection results. We rigorously explore the impact of each technique on detection performance. Experimental results demonstrate significant improvements in detection accuracy. Our method achieved a top score in the Development category, with a public AP of 0.732 and a private AP of 27.2, both at IoU=0.5.