연세대학교 ‘데이터베이스시스템응용’ 수업에서 진행한 강의 자료입니다.

비디오 이상 탐지(Video Anomaly Detection) 주제에 대한 이론과 실습 내용을 다룹니다.

실습 코드는 아래의 실습 pdf에서 확인해보실 수 있습니다.

[이론]
1. Deep One-Class Classification (ICML, 2018)
2. Unsupervised anomaly detection with generative adversarial networks to guide marker discovery (IPMI, 2017)
3. Reconstruction by inpainting for visual anomaly detection (Pattern Recognition, 2021)
4. Cutpaste: Self-supervised learning for anomaly detection and localization (CVPR, 2021)
5. Learning Temporal Regularity in Video Sequences (CVPR, 2016)
6. Future Frame Prediction for Anomaly Detection – A New Baseline (CVPR, 2018)
7. Real-world anomaly detection in surveillance videos (CVPR, 2018)
8. Generative Cooperative Learning for Unsupervised Video Anomaly Detection(CVPR, 2022)
9. Video Anomaly Detection by Solving Decoupled Spatio-Temporal Jigsaw Puzzles (ECCV, 2022)
10. Feature Prediction Diffusion Model for Video Anomaly Detection (ICCV, 2023)
11. Multimodal Motion Conditioned Diffusion Model (CVPR, 2023)
12. TEVAD: Improved video anomaly detection with captions (CVPR, 2023)

[실습]
1. Future Frame Prediction for Anomaly Detection – A New Baseline (CVPR, 2018)

[제작]
연세대학교 컴퓨터과학과 DELAB – Multi Modal Deep Learning Team (24.04.18)

이론

<PT 자료 – PDF>

실습

<PT 자료 – PDF>