Retrieval in Long-Surveillance Videos Using User-Described Motion and Object Attributes

Abstract

Abstract-We present a content-based retrieval method for long surveillance videos both for wide-area (Airborne) as well as near-field imagery (CCTV). Our goal is to retrieve video segments, with a focus on detecting objects moving on routes, that match user-defined events of interest. The sheer size and remote locations where surveillance videos are acquired, necessitates highly compressed representations that are also meaningful for supporting user-defined queries. To address these challenges we archive long-surveillance video through lightweight processing based on low-level local spatio-temporal extraction of motion and object features. These are then hashed into an inverted index using locality-sensitive hashing (LSH). This local approach allows for query flexibility as well as leads to significant gains in compression. Our second task is to extract partial matches to the user-created query and assembles them into full matches using Dynamic Programming (DP). DP exploits causality to assemble the indexed low level features into a video segment which matches the query route. We examine CCTV and Airborne footage, whose low contrast makes motion extraction more difficult. We generate robust motion estimates for Airborne data using a tracklets generation algorithm while we use Horn and Schunck approach to generate motion estimates for CCTV. Our approach handles long routes, low contrasts and occlusion. We derive bounds on the rate of false positives and demonstrate the effectiveness of the approach for counting, motion pattern recognition and abandoned object applications.

Dont Even Look Once: Synthesizing Features for Zero-Shot Detection

“…Semantic information is a key ingredient for transferring knowledge from seen to unseen classes. This can be in the form of attributes [10,21,28,29,2,5], word phrases [38,11], etc. Such semantic data is often easier to collect and the premise of many ZSL methods is to substitute hard to collect visual samples for semantic data.…”

Sand Dust Images Enhancement Based on Red and Blue Channels

“…Therefore, these degraded sand dust images will greatly lose their quality and degrade the performance of computer vision application systems that typically work outdoors during inclement weather conditions. Such systems include video surveillance systems for public security monitoring [1,2], intelligent transportation systems for license plate recognition [3,4], visual recognition systems for automatic driving [5], and so on. Hence, developing an effective sand dust image restoration method to restore color and contrast for computer vision application systems is desirable.…”

A Systematic Review of Intelligence Video Surveillance: Trends, Techniques, Frameworks, and Datasets

“…The implementation of this approach can, therefore, help the system to give real-time alert and forensic examination results for the security staffs because it is supported by an advanced video analysis technique. The implementation of advanced use of video surveillance system in recent decades has led to various domains, such as crime prevention [2], [3]; elder treatment [4]; accident detection [5], [6] traffic oversee and control [5], [7]- [11]; counting moving object such as pedestrians or vehicles [12]- [14]; human activity understanding [15]- [22]; motion detection [23]- [27], activity analysis [16], [28]- [30], identifying, tracking, and classifying vehicles, human, and any object of interest [3], [31]- [37]. The availability, development, and price of processors and sensors have also led to the utilization of this system in term of supporting indoor and outdoor neighborhoods such as shopping mall, airport, train station, and parking lots [16], [20], [38]- [40].…”