Macarulla Rodriguez Andrea, Geradts Zeno, Worring Marcel, Unzueta Luis
Netherlands Forensic Institute, Laan van Ypenburg 6, The Hague, 2497GB, the Netherlands.
University of Amsterdam, Science Park 904, Amsterdam, 1098XH, the Netherlands.
Forensic Sci Int Synerg. 2024 Feb 29;8:100458. doi: 10.1016/j.fsisyn.2024.100458. eCollection 2024.
In forensic and security scenarios, accurate facial recognition in surveillance videos, often challenged by variations in pose, illumination, and expression, is essential. Traditional manual comparison methods lack standardization, revealing a critical gap in evidence reliability. We propose an enhanced images-to-video recognition approach, pairing facial images with attributes like pose and quality. Utilizing datasets such as ENFSI 2015, SCFace, XQLFW, ChokePoint, and ForenFace, we assess evidence strength using calibration methods for likelihood ratio estimation. Three models-ArcFace, FaceNet, and QMagFace-undergo validation, with the log-likelihood ratio cost () as a key metric. Results indicate that prioritizing high-quality frames and aligning attributes with reference images optimizes recognition, yielding similar values to the top 25% best frames approach. A combined embedding weighted by frame quality emerges as the second-best method. Upon preprocessing facial images with the super resolution CodeFormer, it unexpectedly increased , undermining evidence reliability, advising against its use in such forensic applications.
在法医和安全场景中，监控视频中的准确人脸识别至关重要，但其常受到姿势、光照和表情变化的挑战。传统的人工比对方法缺乏标准化，这在证据可靠性方面暴露出关键差距。我们提出了一种增强的图像到视频识别方法，将面部图像与姿势和质量等属性配对。利用ENFSI 2015、SCFace、XQLFW、ChokePoint和ForenFace等数据集，我们使用似然比估计的校准方法评估证据强度。对ArcFace、FaceNet和QMagFace这三种模型进行验证，将对数似然比成本（）作为关键指标。结果表明，优先选择高质量帧并将属性与参考图像对齐可优化识别，产生与前25%最佳帧方法相似的值。由帧质量加权的组合嵌入法成为第二好的方法。在用超分辨率CodeFormer对面部图像进行预处理后，意外地增加了，这有损证据可靠性，建议不要在这类法医应用中使用。
