Please use this identifier to cite or link to this item:
https://elibrary.tucl.edu.np/handle/123456789/18802
Title: | ANOMALY DETECTION IN SURVILLENCE VIDEOS |
Authors: | GURAGAIN, JIWAN PRASAD SHRESTHA, KUSHAL KUNWAR, LAXMAN SUBEDI, YAMAN |
Keywords: | Computer Vision,;Deep Learning,;Video Processing |
Issue Date: | May-2023 |
Publisher: | I.O.E. Pulchowk Campus |
Institute Name: | Institute of Engineering |
Level: | Bachelor |
Abstract: | Anomaly detection with weakly supervised video-level labels is typically formulated as a multiple instance learning (MIL) problem, in which we aim to identify snippets containing abnormal events, with each video represented as a bag of video snippets. Although current methods show effective detection performance, their recognition of the positive instances, i.e., rare abnormal snippets in the abnormal videos, is largely biased by the dominant negative instances, especially when the abnormal events are subtle anomalies that exhibit only small differences compared with normal events. This issue is exacerbated in many methods that ignore important video temporal dependencies. To address this issue, we use add information from the optical flow which captures the temporal relation between successive frames in a video. In this project, we explored the field of video anomaly detection and reviewed existing literature on the subject, as well as related topics such as action recognition and optical flow extraction. |
Description: | Anomaly detection with weakly supervised video-level labels is typically formulated as a multiple instance learning (MIL) problem, in which we aim to identify snippets containing abnormal events, with each video represented as a bag of video snippets. |
URI: | https://elibrary.tucl.edu.np/handle/123456789/18802 |
Appears in Collections: | Computer Engineering |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
Jiwan pd Guragain et al. be project report electronics and computer eng may 2023.pdf | 11.94 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.