Weakly supervised marine animal detection from remote sensing images using vector-quantized variational autoencoder

TL;DR

This paper introduces a weakly-supervised marine animal detection method using vector-quantized variational autoencoders, improving anomaly localization and interpretability in aerial marine imagery, with superior results over existing techniques.

Contribution

It adapts vector-quantized variational autoencoders for marine animal detection, addressing noisy data and enhancing anomaly interpretability in aerial images.

Findings

Outperforms recent methods in marine animal detection

Provides improved anomaly localization and interpretability

Demonstrates effectiveness on two dedicated datasets

Abstract

This paper studies a reconstruction-based approach for weakly-supervised animal detection from aerial images in marine environments. Such an approach leverages an anomaly detection framework that computes metrics directly on the input space, enhancing interpretability and anomaly localization compared to feature embedding methods. Building upon the success of Vector-Quantized Variational Autoencoders in anomaly detection on computer vision datasets, we adapt them to the marine animal detection domain and address the challenge of handling noisy data. To evaluate our approach, we compare it with existing methods in the context of marine animal detection from aerial image data. Experiments conducted on two dedicated datasets demonstrate the superior performance of the proposed method over recent studies in the literature. Our framework offers improved interpretability and localization of anomalies, providing valuable insights for monitoring marine ecosystems and mitigating the impact of human activities on marine animals.