Authors:
Sai Vishwanath Venkatesh
1
;
Adithya Prem Anand
2
;
Gokul Sahar S.
2
;
Akshay Ramakrishnan
2
and
Vineeth Vijayaraghavan
3
Affiliations:
1
SRM, Institute of Science and Technology, Chennai, India
;
2
SSN, College of Engineering, Chennai, Tamil Nadu, India
;
3
Solarillion Foundation, Chennai, Tamil Nadu, India
Keyword(s):
Real-time, Surveillance, Edge Devices, Resource-constrained, Crime Detection.
Abstract:
There is a growing use of surveillance cameras to maintain a log of events that would help in the identification of criminal activities. However, it is necessary to continuously monitor the acquired footage which contributes to increased labor costs but more importantly, violation of privacy. Therefore, we need decentralized surveillance systems that function autonomously in real-time to reduce crime rates even further. In our work, we discuss an efficient method of crime detection using Deep Learning, that can be used for on-device crime monitoring. By making the inferences on-device, we can reduce the latency, the cost overhead for the collection of data into a centralized unit and counteract the lack of privacy. Using the concept of EarlyStopping–Multiple Instance Learning to provide low inference time, we build specialized models for crime detection using two real-world datasets in the domain. We implement the concept of Sub-Nyquist sampling on a video and introduce a metric ηcom
p for evaluating the reduction of computation due to undersampling. On average, our models tested on Raspberry Pi 3 Model B provide a 30% increase of accuracy over benchmarks, computational savings as 80.23% and around 13 times lesser inference times. This allows for the development of efficient and accurate real-time implementation on edge devices for on-device crime detection.
(More)