arXiv 2505.046162025-05-07cs.CV

高空与远距离下的人员识别：人脸、体型与步态的融合

Person Recognition at Altitude and Range: Fusion of Face, Body Shape and Gait

Department of Computer Science and Engineering, Michigan State University（计算机科学与工程系，密歇根州立大学）；School of Electrical and Computer Engineering, Purdue University（电气与计算机工程学院，普渡大学）；School of Interactive Computing, Georgia Tech（交互计算学院，佐治亚理工学院）；Department of Electrical and Computer Engineering, University of Texas at Austin（电气与计算机工程系，德克萨斯大学奥斯汀分校）

Feng Liu, Nicholas Chimitt, Lanqing Guo, Jitesh Jain, Aditya Kane, Minchul Kim, Wes Robbins, Yiyang Su, Dingqiang Ye, Xingguang Zhang, Jie Zhu, Siddharth Satyak… 展开作者

Feng Liu, Nicholas Chimitt, Lanqing Guo, Jitesh Jain, Aditya Kane, Minchul Kim, Wes Robbins, Yiyang Su, Dingqiang Ye, Xingguang Zhang, Jie Zhu, Siddharth Satyakam, Christopher Perry, Stanley H. Chan, Arun Ross, Humphrey Shi, Zhangyang Wang, Anil Jain, Xiaoming Liu

备注：: 18 pages, 12 figures

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文提出 FarSight 端到端系统，融合人脸、体型与步态，通过检测跟踪、视频恢复、特征编码和质量引导融合，在 BRIAR 远距离识别任务上显著提升验证、闭集与开集识别性能。

摘要：

我们研究无约束环境中的全身人员识别问题。该问题出现在如 IARPA 高空与远距离生物特征识别与鉴定（BRIAR）项目等监控场景中，其中生物特征数据在远距离、高视角以及恶劣大气条件（例如湍流和高风速）下采集。为此，我们提出 FarSight，一个统一的端到端人员识别系统，融合人脸、步态和体型模态中的互补生物特征线索。FarSight 在四个核心模块中引入了新算法：多目标检测与跟踪、识别感知的视频恢复、模态特定的生物特征编码以及质量引导的多模态融合。这些组件被设计为在图像退化、大幅度姿态与尺度变化以及跨域差异下协同工作。在 BRIAR 数据集（用于远距离多模态生物特征识别的最全面基准之一）上的大量实验证明了 FarSight 的有效性。与我们的初步系统相比，该系统在 1:1 验证准确率（TAR@0.1% FAR）上取得了 34.1% 的绝对提升，在闭集识别（Rank-20）上提升了 17.8%，并在开集识别错误（FNIR@1% FPIR）上降低了 34.3%。此外，FarSight 在 2025 年 NIST RTE 视频人脸评估（FIVE）中接受了评估，该评估在 BRIAR 数据集上进行标准化人脸识别测试。这些结果确立了 FarSight 作为在具有挑战性的现实条件下进行可操作生物特征识别的最先进解决方案。

英文摘要：

We address the problem of whole-body person recognition in unconstrained environments. This problem arises in surveillance scenarios such as those in the IARPA Biometric Recognition and Identification at Altitude and Range (BRIAR) program, where biometric data is captured at long standoff distances, elevated viewing angles, and under adverse atmospheric conditions (e.g., turbulence and high wind velocity). To this end, we propose FarSight, a unified end-to-end system for person recognition that integrates complementary biometric cues across face, gait, and body shape modalities. FarSight incorporates novel algorithms across four core modules: multi-subject detection and tracking, recognition-aware video restoration, modality-specific biometric feature encoding, and quality-guided multi-modal fusion. These components are designed to work cohesively under degraded image conditions, large pose and scale variations, and cross-domain gaps. Extensive experiments on the BRIAR dataset, one of the most comprehensive benchmarks for long-range, multi-modal biometric recognition, demonstrate the effectiveness of FarSight. Compared to our preliminary system, this system achieves a 34.1% absolute gain in 1:1 verification accuracy (TAR@0.1% FAR), a 17.8% increase in closed-set identification (Rank-20), and a 34.3% reduction in open-set identification errors (FNIR@1% FPIR). Furthermore, FarSight was evaluated in the 2025 NIST RTE Face in Video Evaluation (FIVE), which conducts standardized face recognition testing on the BRIAR dataset. These results establish FarSight as a state-of-the-art solution for operational biometric recognition in challenging real-world conditions.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。