Xavier Francis | Searcharxiv

arXiv2021

A comparison of low-cost behavioral observation software applications and recommendations for use

In the field of animal behavior and behavioral ecology, many standardized methods to observe animal behavior were established in the last decades. While the protocols remained similar, behavioral researchers can take advantage of technological advancements to enter observations directly onto a handheld computer (phone, tablet, etc.), saving time and potentially increasing fidelity of recordings. However, we now have the choice between many different platforms for recording behavioral observations. Our challenge is choosing the most appropriate platform that fits a particular study question, research design, budget, and desired amount of preparatory time. Here, we review six low-cost software applications for handheld computers that are available for real-time entry of behavioral observations: Animal Behaviour Pro, Animal Observer, BORIS, CyberTracker, Prim8, and ZooMonitor. We discuss the preliminary decisions that have to be made about the study design, and we assess the six applications by providing the advantages and disadvantages of each platform and an overall application comparison. Our goal is to help researchers make calculated decisions about what behavioral observation platform is best for their study system and question.

cs.SI↗

arXiv2019

Learning Wear Patterns on Footwear Outsoles Using Convolutional Neural Networks

Footwear outsoles acquire characteristics unique to the individual wearing them over time. Forensic scientists largely rely on their skills and knowledge, gained through years of experience, to analyse such characteristics on a shoeprint. In this work, we present a convolutional neural network model that can predict the wear pattern on a unique dataset of shoeprints that captures the life and wear of a pair of shoes. We present an additional architecture able to reconstruct the outsole back to its original state on a given week, and provide empirical evaluations of the performance of both models.

cs.CV↗