LA Referencia (Red Federada de Repositorios Institucionales de Publicaciones Científicas) · 2021 · 13 citations · 39 references
Open access
ComputingComputer ScienceCriminologyComputer VisionCrimeMachine LearningAction RecognitionCrime PreventionDeep LearningPattern RecognitionVideo AnalysisVisual TrackingAction DetectionActivity RecognitionCrime Pattern DetectionIntention DetectionSuspicious Behavior DetectionVideo Action RecognitionVideo Anomaly DetectionVideo Event Detection
Crime generates significant losses, both human and economic. Every year, billions of\n\t\t\t\t dollars are lost due to attacks, crimes, and scams. Surveillance video camera networks generate\n\t\t\t\t vast amounts of data, and the surveillance staff cannot process all the information in real-time.\n\t\t\t\t Human sight has critical limitations. Among those limitations, visual focus is one of the most\n\t\t\t\t critical when dealing with surveillance. For example, in a surveillance room, a crime can occur\n\t\t\t\t in a different screen segment or on a distinct monitor, and the surveillance staff may overlook it.\n\t\t\t\t Our proposal focuses on shoplifting crimes by analyzing situations that an average person will\n\t\t\t\t consider as typical conditions, but may eventually lead to a crime. While other approaches identify\n\t\t\t\t the crime itself, we instead model suspicious behavior—the one that may occur before the build-up\n\t\t\t\t phase of a crime—by detecting precise segments of a video with a high probability of containing\n\t\t\t\t a shoplifting crime. By doing so, we provide the staff with more opportunities to act and prevent\n\t\t\t\t crime. We implemented a 3DCNN model as a video feature extractor and tested its performance\n\t\t\t\t on a dataset composed of daily action and shoplifting samples. The results are encouraging as\n\t\t\t\t the model correctly classifies suspicious behavior in most of the scenarios where it was tested.\n\t\t\t\t For example, when classifying suspicious behavior, the best model generated in this work obtains\n\t\t\t\t precision and recall values of 0.8571 and 1 in one of the test scenarios, respectively.
39
Deep Residual Learning for Image Recognition
Kaiming He, Xiangyu Zhang, Shaoqing Ren et al. · 2016 · 222.1K citations · Full text
Going deeper with convolutions
Christian Szegedy, Wei Liu, Yangqing Jia et al. · 2015 · 46.8K citations
Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning
Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke et al. · 2017 · 12.6K citations · Full text