Learnable Locality-Sensitive Hashing for Video Anomaly Detection

TL;DR

This paper introduces a learnable, neural network-embedded locality-sensitive hashing method for video anomaly detection, improving scalability, robustness, and accuracy over existing reconstruction-based approaches.

Contribution

It proposes a novel learnable LSH technique embedded in a neural network, optimized with contrastive learning for effective video anomaly detection.

Findings

Achieves state-of-the-art results on VAD benchmarks.

Robust to data imbalance and intra-class variations.

Scalable and efficient for large datasets.

Abstract

Video anomaly detection (VAD) mainly refers to identifying anomalous events that have not occurred in the training set where only normal samples are available. Existing works usually formulate VAD as a reconstruction or prediction problem. However, the adaptability and scalability of these methods are limited. In this paper, we propose a novel distance-based VAD method to take advantage of all the available normal data efficiently and flexibly. In our method, the smaller the distance between a testing sample and normal samples, the higher the probability that the testing sample is normal. Specifically, we propose to use locality-sensitive hashing (LSH) to map samples whose similarity exceeds a certain threshold into the same bucket in advance. In this manner, the complexity of near neighbor search is cut down significantly. To make the samples that are semantically similar get closer and samples not similar get further apart, we propose a novel learnable version of LSH that embeds LSH into a neural network and optimizes the hash functions with contrastive learning strategy. The proposed method is robust to data imbalance and can handle the large intra-class variations in normal data flexibly. Besides, it has a good ability of scalability. Extensive experiments demonstrate the superiority of our method, which achieves new state-of-the-art results on VAD benchmarks.