2022/10/29 by Xiaochen Zheng, Zheng, Xiaochen
Biochemistry, Genetics and Molecular Biology · Computer Science · Environmental Science · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #Identification and Quantification in Food #Wildlife Ecology and Conservation
paper · pdf · doi:10.48550/arxiv.2211.05636
openalex publication_date 2022/10/29 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
Automated animal censuses with aerial imagery are a vital ingredient towards wildlife conservation. Recent models are generally based on supervised learning and thus require vast amounts of training data. Due to their scarcity and minuscule size, annotating animals in aerial imagery is a highly tedious process. In this project, we present a methodology to reduce the amount of required training data by resorting to self-supervised pretraining. In detail, we examine a combination of recent contrastive learning methodologies like Momentum Contrast (MoCo) and Cross-Level Instance-Group Discrimination (CLD) to condition our model on the aerial images without the requirement for labels. We show that a combination of MoCo, CLD, and geometric augmentations outperforms conventional models pretrained on ImageNet by a large margin. Meanwhile, strategies for smoothing label or prediction distribution in supervised learning have been proven useful in preventing the model from overfitting. We combine the self-supervised contrastive models with image mixup strategies and find that it is useful for learning more robust visual representations. Crucially, our methods still yield favorable results even if we reduce the number of training animals to just 10%, at which point our best model scores double the recall of the baseline at similar precision. This effectively allows reducing the number of required annotations to a fraction while still being able to train high-accuracy models in such highly challenging settings.