Cell Phone Recognition
Cell phone recognition research aims to accurately identify the specific device used to record audio or visual data. Current efforts focus on improving recognition accuracy for flexible objects (like cell phones) by developing models that better handle variations in shape, size, and lighting conditions, often employing graph neural networks or evidential modeling to quantify uncertainty and improve robustness. These advancements are significant for forensic science, security applications, and improving the reliability of multimedia data analysis by enabling more precise source identification.
Papers
June 6, 2024
September 14, 2023