Skip to content

Cross-Domain Image Matching with Deep Feature Maps

Journal: International Journal of Computer Vision
Published: 2019
Primary Author: Bailey Kong
Secondary Authors: James Supancic, Deva Ramana, Charless Fowlkes
Research Area: Footwear

We investigate the problem of automatically determining what type of shoe left an impression found at a crime scene. This recognition problem is made difficult by the variability in types of crime scene evidence (ranging from traces of dust or oil on hard surfaces to impressions made in soil) and the lack of comprehensive databases of shoe outsole tread patterns. We find that mid-level features extracted by pre-trained convolutional neural nets are surprisingly effective descriptors for this specialized domains. However, the choice of similarity measure for matching exemplars to a query image is essential to good performance. For matching multi-channel deep features, we propose the use of multi-channel normalized cross-correlation and analyze its effectiveness. Our proposed metric significantly improves performance in matching crime scene shoeprints to laboratory test impressions. We also show its effectiveness in other cross-domain image retrieval problems: matching facade images to segmentation labels and aerial photos to map images. Finally, we introduce a discriminatively trained variant and fine-tune our system through our proposed metric, obtaining state-of-the-art performance.

Related Resources

Assessing Footwear Examiner Performance

Assessing Footwear Examiner Performance

This CSAFE webinar was held on May 13, 2021. Presenter: Corey Katz Graduate Researcher, University of California, Irvine Presentation Description: There has recently been increased attention on the reliability and…
Open Forensic Science in R

Open Forensic Science in R

This book is for anyone looking to do forensic science analysis in a data-driven and open way. Whether you are a student, teacher, or scientist, this book is for you. We take…
Quantifying the similarity of 2D images using edge pixels: An application to the forensic comparison of footwear impressions

Quantifying the similarity of 2D images using edge pixels: An application to the forensic comparison of footwear impressions

We propose a novel method to quantify the similarity between an impression (Q) from an unknown source and a test impression (K) from a known source. Using the property of…
Quantifying the similarity of 2D images using edge pixels: An application to the forensic comparison of footwear impressions

Quantifying the similarity of 2D images using edge pixels: An application to the forensic comparison of footwear impressions

We propose a novel method to quantify the similarity between an impression (Q) from an unknown source and a test impression (K) from a known source. Using the property of…