Unsupervised data augmentation for object detection

TL;DR

This paper introduces an unsupervised data augmentation framework for object detection using GANs, enabling the generation of labeled images with objects in specific positions to improve model training.

Contribution

It presents a novel two-step GAN-based pipeline that generates labeled images with objects in desired locations, addressing the challenge of augmentation in object detection.

Findings

Enhanced data diversity for object detection models.

Generated images with accurate bounding boxes for training.

Potential improvements in detection accuracy.

Abstract

Data augmentation has always been an effective way to overcome overfitting issue when the dataset is small. There are already lots of augmentation operations such as horizontal flip, random crop or even Mixup. However, unlike image classification task, we cannot simply perform these operations for object detection task because of the lack of labeled bounding boxes information for corresponding generated images. To address this challenge, we propose a framework making use of Generative Adversarial Networks(GAN) to perform unsupervised data augmentation. To be specific, based on the recently supreme performance of YOLOv4, we propose a two-step pipeline that enables us to generate an image where the object lies in a certain position. In this way, we can accomplish the goal that generating an image with bounding box label.