Progressive Transformation Learning for Leveraging Virtual Images in Training

TL;DR

This paper introduces Progressive Transformation Learning (PTL), a method that iteratively enhances virtual images' realism to improve UAV-based object detection, especially in small data and cross-domain scenarios.

Contribution

The paper proposes a novel PTL framework that progressively augments training data with transformed virtual images, effectively reducing domain gap issues in UAV object detection.

Findings

PTL significantly improves detection performance over baseline methods.

PTL is especially effective in small data and cross-domain settings.

The Mahalanobis distance effectively quantifies domain gap for image selection.

Abstract

To effectively interrogate UAV-based images for detecting objects of interest, such as humans, it is essential to acquire large-scale UAV-based datasets that include human instances with various poses captured from widely varying viewing angles. As a viable alternative to laborious and costly data curation, we introduce Progressive Transformation Learning (PTL), which gradually augments a training dataset by adding transformed virtual images with enhanced realism. Generally, a virtual2real transformation generator in the conditional GAN framework suffers from quality degradation when a large domain gap exists between real and virtual images. To deal with the domain gap, PTL takes a novel approach that progressively iterates the following three steps: 1) select a subset from a pool of virtual images according to the domain gap, 2) transform the selected virtual images to enhance realism, and 3) add the transformed virtual images to the training set while removing them from the pool. In PTL, accurately quantifying the domain gap is critical. To do that, we theoretically demonstrate that the feature representation space of a given object detector can be modeled as a multivariate Gaussian distribution from which the Mahalanobis distance between a virtual object and the Gaussian distribution of each object category in the representation space can be readily computed. Experiments show that PTL results in a substantial performance increase over the baseline, especially in the small data and the cross-domain regime.