A case study in identifying acceptable bitrates for human face recognition tasks

作者：

Highlights：

• Identification of acceptable bitrates for human face recognition systems.

• The London buses application is the case study for the applied methodology.

• Results show the influence of scene content on acceptance thresholds.

• Dark scenes and scenes with high spatial-temporal busyness require lower compression.

摘要

•Identification of acceptable bitrates for human face recognition systems.•The London buses application is the case study for the applied methodology.•Results show the influence of scene content on acceptance thresholds.•Dark scenes and scenes with high spatial-temporal busyness require lower compression.

论文关键词：CCTV recording systems,Human face recognition,H.264/MPEG-4 AVC,Image usefulness,Visual psychophysics,Scene characterization

论文评审过程：Received 19 September 2014, Revised 27 January 2015, Accepted 7 May 2015, Available online 15 May 2015, Version of Record 15 June 2015.

论文官网地址：https://doi.org/10.1016/j.image.2015.05.002