Saverio Cavasin, Pietro Tedeschi, Mattia Tamiazzo, Alessandro Brighente, Simone Milani, Mauro Conti
The VESSI framework enhances automated video surveillance analysis using Vision-Language Models (VLMs) to convert visual features into textual descriptions. This improves situational awareness and supports post-incident investigations. VESSI substantially improves analysis capabilities and reduces review time.
The VESSI framework has practical significance in intelligence infrastructures and Law Enforcement agencies, where automated video surveillance analysis is critical. It can reduce the burden on human analysts and improve operational insight. This can lead to more efficient and effective surveillance and investigation processes.
Thank you to arXiv for use of its open access interoperability. arxiv.org. aibrief.ai is not affiliated with or endorsed by arXiv.