実世界の異常分類に向けた潜在空間クラスタの条件付け
Conditioning Latent-Space Clusters for Real-World Anomaly Classification
変分オートエンコーダの潜在空間を条件付けし、都市景観の高解像度カメラ画像から正常・異常を分類する手法を提案。小さな異常に対応するため不一致マップを追加入力し、高品質な再構成を保ちつつ潜在表現を分離。
著者: Daniel Bogdoll, Svetlana Pavlitska, Simon Klaus, J. Marius Zöllner
分類: cs.CV, cs.AI, cs.RO
原文アブストラクト
Anomalies in the domain of autonomous driving are a major hindrance to the large-scale deployment of autonomous vehicles. In this work, we focus on high-resolution camera data from urban scenes that include anomalies of various types and sizes. Based on a Variational Autoencoder, we condition its latent space to classify samples as either normal data or anomalies. In order to emphasize especially small anomalies, we perform experiments where we provide the VAE with a discrepancy map as an additional input, evaluating its impact on the detection performance. Our method separates normal data and anomalies into isolated clusters while still reconstructing high-quality images, leading to meaningful latent representations.
