Shoeprints and Tread Marks

CSAFE researchers are investigating spatial distribution and uniqueness of tread pattern features of shoe outsoles using a large database of shoe/brand types. Researchers are also making advancements in the development of 3D shoeprint scanning technology. Expansions in the use of this new technology will transform the process of shoeprint database construction and modeling, allowing for increased confidence in the determination that a crime scene shoeprint matches a putative shoe.

Area

Pattern Evidence

Disciplines

Statistics

Research

  • Assessing independence and uniformity assumptions currently in use for quantifying the strength of shoeprint evidence based on acquired characteristics, developing new, more appropriate probability models for quantifying uncertainty in matching shoes to prints using accidental/randomly acquired characteristics (RACs), and investigating how accidental/randomly acquired characteristics should be represented and compared during matching, which will influence the development of probability model.
  • Creating a longitudinal database of similar shoes that can be used for research and black-box experiments, defining a score to characterize a shoeprint using 2D images that combine global and local attributes, analyzing wear, and persistence and evolution of RACs through time, and more.

Benefits of Research

CSAFE is helping to pave the way for greater objectivity when analyzing shoeprints and tread marks at a scene of a crime. CSAFE shoeprint research has the potential to provide an intuitive, rigorous and statistically sound basis for expert forensic testimony related to shoeprint evidence. The statistical model and computational algorithms could be applicable to a wider range of problems.

Select Publications, Conference Papers, Presentations and/or Tools

Bailey Kong, James Supancic, Deva Ramanan, Charless Fowlkes, “Cross-Domain Forensic Shoeprint Matching”, Proceedings of British Machine Vision Conference, London, 2017. (Awarded an honorable mention for “Best Industrial Paper”)

Charless Fowlkes, Bailey Kong, James Supancic, Deva Ramanan, “Learning features for matching class characteristics in footwear impression evidence”, 10th International Conference on Forensic Inference and Statistics, Minneapolis, MN, Sept. 2017

Poster presentation, “Shoe matching based on image cross-sectioning,” 10th International Conference on Forensic Inference and Statistics, Minneapolis, MN, September, 2017.

Poster presentation, “Incorporating a Statistical Model into Forensic Shoeprint Analysis ,” 70th Annual Scientific Meeting of the American Academy of Forensic Sciences, Seattle, WA, February 2018.