Search Results for author: Keval Doshi

Found 13 papers, 4 papers with code

Adversarial Machine Learning Attacks Against Video Anomaly Detection Systems

no code implementations7 Apr 2022 Furkan Mumcu, Keval Doshi, Yasin Yilmaz

We demonstrate how Wi-Fi deauthentication attack, a notoriously easy-to-perform and effective denial-of-service (DoS) attack, can be utilized to generate adversarial data for video anomaly detection systems.

Anomaly Detection BIG-bench Machine Learning +1

TiSAT: Time Series Anomaly Transformer

1 code implementation10 Mar 2022 Keval Doshi, Shatha Abudalou, Yasin Yilmaz

While anomaly detection in time series has been an active area of research for several years, most recent approaches employ an inadequate evaluation criterion leading to an inflated F1 score.

Anomaly Detection Time Series

Zero-Shot Action Recognition with Transformer-based Video Semantic Embedding

no code implementations10 Mar 2022 Keval Doshi, Yasin Yilmaz

While video action recognition has been an active area of research for several years, zero-shot action recognition has only recently started gaining traction.

Action Recognition Zero-Shot Action Recognition

Rethinking Video Anomaly Detection - A Continual Learning Approach

no code implementations WACV 2022 Keval Doshi, Yasin Yilmaz

The experimental results show that the existing state-of-the-art methods are not suitable for the considered practical challenges, and the proposed algorithm outperforms them with a large margin in continual learning and few-shot learning tasks

Anomaly Detection Continual Learning +1

An Efficient Approach for Anomaly Detection in Traffic Videos

no code implementations20 Apr 2021 Keval Doshi, Yasin Yilmaz

We also propose a sequential change detection algorithm that can quickly adapt to a new scene and detect changes in the similarity statistic.

Anomaly Detection Change Detection

Road Damage Detection using Deep Ensemble Learning

1 code implementation30 Oct 2020 Keval Doshi, Yasin Yilmaz

Road damage detection is critical for the maintenance of a road, which traditionally has been performed using expensive high-performance sensors.

Ensemble Learning Management +1

Online Anomaly Detection in Surveillance Videos with Asymptotic Bounds on False Alarm Rate

1 code implementation10 Oct 2020 Keval Doshi, Yasin Yilmaz

Motivated by these research gaps, we propose an online anomaly detection method in surveillance videos with asymptotic bounds on the false alarm rate, which in turn provides a clear procedure for selecting a proper decision threshold that satisfies the desired false alarm rate.

Anomaly Detection In Surveillance Videos Decision Making

Fast Unsupervised Anomaly Detection in Traffic Videos

1 code implementation CVPR 2020 Keval Doshi, Yasin Yilmaz

In this paper, we propose a fast unsupervised anomaly detection system comprising of three modules: preprocessing module, candidate selection module and backtracking anomaly detection module.

Unsupervised Anomaly Detection

Timely Detection and Mitigation of Stealthy DDoS Attacks via IoT Networks

no code implementations15 Jun 2020 Keval Doshi, Yasin Yilmaz, Suleyman Uludag

Internet of Things (IoT) networks consist of sensors, actuators, mobile and wearable devices that can connect to the Internet.

Intrusion Detection

Any-Shot Sequential Anomaly Detection in Surveillance Videos

no code implementations5 Apr 2020 Keval Doshi, Yasin Yilmaz

Even though the performance of state-of-the-art methods on publicly available data sets has been competitive, they demand a massive amount of training data.

Anomaly Detection In Surveillance Videos Decision Making +1

Synthetic Image Augmentation for Improved Classification using Generative Adversarial Networks

no code implementations31 Jul 2019 Keval Doshi

Object detection and recognition has been an ongoing research topic for a long time in the field of computer vision.

General Classification Image Augmentation +2

Cannot find the paper you are looking for? You can Submit a new open access paper.