Get Started
Home
Topics
Search
Library
Research questionHow does facial-search gallery size affect witness accuracy and confidence in photo lineups?A facial-search system can place a top-ranked gallery image into a lineup even when it is not the probe subject. This makes a witness’s subsequent choice and confidence harder to interpret in an arrest decision.
AI
Computer Vision
Image & Video Processing
Research Paper
Latest papersRecent research connected to this question, newest first.What Happens to Accuracy When Photo Lineups Contain Non-Mated Rank-One Images From Large Galleries?The evidence concerns one-to-many facial identification using surveillance-video probes against driver’s-license and/or booking-photo galleries, followed by a witness photo lineup. It compares galleries of 500, 5,000, and 24,000 images, using an algorithm’s rank-one image or a human examiner’s selection from top-ranked results, and reports more incorrect identifications and higher confidence in those errors with larger galleries.research paper · Sep 2, 2026
Related questions
Can multimodal models match human judgments of facial attractiveness, not merely rank faces correctly?How can video facial-expression recognition personalize vision-language models under shifts without costly test-time optimization?How can document retrieval find the correct record when large collections share nearly identical visual templates?How can post-hoc vision saliency remain faithful when image orientation changes?