“Forget” the Forget Gate: Estimating Anomalies in Videos Using Self-contained Long Short-Term Memory Networks

Habtamu Fanta, Zhiwen Shao, Lizhuang Ma

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

1 Scopus citations

Abstract

Abnormal event detection is a challenging task that requires effectively handling intricate features of appearance and motion. In this paper, we present an approach of detecting anomalies in videos by learning a novel LSTM based self-contained network on normal dense optical flow. Due to their sigmoid implementations, standard LSTM’s forget gate is susceptible to overlooking and dismissing relevant content in long sequence tasks. The forget gate mitigates participation of previous hidden state for computation of cell state prioritizing current input. Besides, the hyperbolic tangent activation of standard LSTMs sacrifices performance when a network gets deeper. To tackle these two limitations, we introduce a bi-gated, light LSTM cell by discarding the forget gate and introducing sigmoid activation. Specifically, the proposed LSTM architecture fully sustains content from previous hidden state thereby enabling the trained model to be robust and make context-independent decision during evaluation. Removing the forget gate results in a simplified and undemanding LSTM cell with improved performance and computational efficiency. Empirical evaluations show that the proposed bi-gated LSTM based network outperforms various LSTM based models for abnormality detection and generalization tasks on CUHK Avenue and UCSD datasets.

Original languageEnglish
Title of host publicationAdvances in Computer Graphics - 37th Computer Graphics International Conference, CGI 2020, Proceedings
EditorsNadia Magnenat-Thalmann, Constantine Stephanidis, George Papagiannakis, Enhua Wu, Daniel Thalmann, Bin Sheng, Jinman Kim, Marina Gavrilova
PublisherSpringer Science and Business Media Deutschland GmbH
Pages169-181
Number of pages13
ISBN (Print)9783030618636
DOIs
StatePublished - 2020
Event37th Computer Graphics International Conference, CGI 2020 - Geneva, Switzerland
Duration: 20 Oct 202023 Oct 2020

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume12221 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Conference

Conference37th Computer Graphics International Conference, CGI 2020
Country/TerritorySwitzerland
CityGeneva
Period20/10/2023/10/20

Keywords

  • Abnormal event detection
  • Abnormality generalization
  • Long Short-Term Memory
  • Self-contained LSTM

Fingerprint

Dive into the research topics of '“Forget” the Forget Gate: Estimating Anomalies in Videos Using Self-contained Long Short-Term Memory Networks'. Together they form a unique fingerprint.

Cite this