Analysing comparative soft biometrics from crowdsourced annotations

• Published in: IET biometrics Vol. 5; no. 4; pp. 276 - 283
• Main Authors: Martinho-Corbishley, Daniel, Nixon, Mark S, Carter, John N
• Format: Journal Article
• Language: English
• Published: Stevenage The Institution of Engineering and Technology 01.12.2016; John Wiley & Sons, Inc
• Subjects: additional discriminative information; Annotations; Biometric identification; Biometrics; biometrics (access control); Cameras; comparative soft biometric analysis; comparative soft biometric body annotations; Correspondence; crowdsourced annotations; Crowdsourcing; Datasets; human description; human identification; Image quality; Image retrieval; Investigations; Labels; low‐quality surveillance footage; pragmatic dataset; RankSVM algorithm; Research Article; Search algorithms; semantic recognition; Semantics; support vector machines; Surveillance; video surveillance; video surveillance operator searching; low-quality surveillance footage; support vector machines; additional discriminative information; semantic recognition; crowdsourced annotations; pragmatic dataset; biometrics (access control); human description; human identification; video surveillance operator searching; RankSVM algorithm; comparative soft biometric analysis; comparative soft biometric body annotations; video surveillance
• ISSN: 2047-4938; 2047-4946; 2047-4946
• DOI: 10.1049/iet-bmt.2015.0118

Soft biometrics enable human description and identification from low-quality surveillance footage. This study premises the design, collection and analysis of a novel crowdsourced dataset of comparative soft biometric body annotations, obtained from a richly diverse set of human annotators. The authors annotate 100 subject images to provide a coherent, in-depth appraisal of the collected annotations and inferred relative labels. The dataset includes gender as a comparative trait and the authors find that comparative labels characteristically contain additional discriminative information over traditional categorical annotations. Using the authors’ pragmatic dataset, semantic recognition is performed by inferring relative biometric signatures using a RankSVM algorithm. This demonstrates a practical scenario, reproducing responses from a video surveillance operator searching for an individual. The approach can reliably return the correct match in the top 7% of results with ten comparisons, or top 13% of results using just five sets of subject comparisons.