Abstract
In this paper we focus on the problem of assembling unbiased photo lineups. Photo lineups are an important method in the identification process for the prosecution and possible conviction of suspects. Incorrect lineup assemblies have led to the false identification and conviction of innocent persons. One of the significant errors which can occur is the lack of lineup fairness, i.e., that the suspect significantly differs from other candidates.
Despite the importance of the task, few tools are available to assist police technicians in creating lineups. Furthermore, these tools mostly focus on fair lineup administration and provide only a little support in the candidate selection process. In our work, we first summarize key personalization and information retrieval (IR) challenges of such systems and propose an IR/Personalization model that addresses them. Afterwards, we describe a LineIT tool that instantiate this model and aims to support police technicians in assembling unbiased lineups.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
The user’s task and therefore expected recommendations may significantly differ between each session. Similarly, for every new lineup, there is a different suspect, which can be considered as a topic drift as well.
- 2.
I.e., we do not aim on “educating” users about the “correct” similarity metric.
- 3.
Currently, a fixed \(3\times 4\) grid with a 10% padding is used, but other variants, e.g., based on the detection of key facial artefacts are plausible.
- 4.
This weight is currently set explicitly via a slider, however, it can be also incorporated in user’s long-term preferences in the future.
- 5.
References
Di Noia, T., Rosati, J., Tomeo, P., Sciascio, E.D.: Adaptive multi-attribute diversity for recommender systems. Inf. Sci. 382(C), 234–253 (2017)
Donahue, J., et al.: DeCAF: a deep convolutional activation feature for generic visual recognition. CoRR abs/1310.1531 (2013). http://arxiv.org/abs/1310.1531
Gallagher, M.: Proportionality, disproportionality and electoral systems. Electoral Stud. 10(1), 33–51 (1991)
Greene, E., Heilbrun, K.: Wrightsman’s Psychology and the Legal System. Cengage Learning, Belmont (2013)
Hidasi, B., Karatzoglou, A.: Recurrent neural networks with top-k gains for session-based recommendations. In: Proceedings of the 27th ACM International Conference on Information and Knowledge Management, CIKM 2018, pp. 843–852. ACM (2018)
Krizhevsky, A., Sutskever, I., Hinton, G.E.: Imagenet classification with deep convolutional neural networks. In: Advances in Neural Information Processing Systems 25, pp. 1097–1105. Curran Associates, Inc. (2012)
MacLin, O.H., Zimmerman, L.A., Malpass, R.S.: Pc\(\_\)eyewitness and the sequential superiority effect: computer-based lineup administration. Law Hum Behav. 29(3), 303–321 (2005)
Mansour, J., Beaudry, J., Kalmet, N., Bertrand, M.I., Lindsay, R.C.L.: Evaluating lineup fairness: variations across methods and measures. Law Hum Behav. 41, 103 (2016)
Parkhi, O.M., Vedaldi, A., Zisserman, A.: Deep face recognition. In: British Machine Vision Conference (2015)
Peska, L., Trojanova, H.: Towards recommender systems for police photo lineup. In: Proceedings of the 2nd Workshop on Deep Learning for Recommender Systems, DLRS 2017, pp. 19–23. ACM (2017)
Peska, L., Trojanova, H.: Personalized recommendations in police photo lineup assembling task. In: Proceedings of the 18th Conference Information Technologies - Applications and Theory (ITAT 2018), No. 2203 in CEUR-WS (2018)
Peska, L., Trojanova, H.: Towards similarity models in police photo lineup assembling tasks. In: Marchand-Maillet, S., Silva, Y.N., Chávez, E. (eds.) SISAP 2018. LNCS, vol. 11223, pp. 217–225. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-02224-2_17
Russakovsky, O., et al.: ImageNet large scale visual recognition challenge. Int. J. Comput. Vis. (IJCV) 115(3), 211–252 (2015)
Skopal, T., Peška, L., Grošup, T.: Interactive product search based on global and local visual-semantic features. In: Marchand-Maillet, S., Silva, Y.N., Chávez, E. (eds.) SISAP 2018. LNCS, vol. 11223, pp. 87–95. Springer, Cham (2018). https://doi.org/10.1007/978-3-030-02224-2_7
Skopal, T., Peška, L., Kovalčík, G., Grosup, T., Lokoč, J.: Product exploration based on latent visual attributes. In: Proceedings of the 2017 ACM on Conference on Information and Knowledge Management, CIKM 2017, pp. 2531–2534. ACM (2017)
Acknowledgements
This work was supported by the grants GAUK-232217, GACR-19-22071Y and 20460-3/2018/FEKUTSTRAT Higher Education Excellence Program 2018. Some resources are available online:
LineIT demo: http://herkules.ms.mff.cuni.cz/lineit_v2,
Source codes: https://github.com/lpeska/LineIT_v2.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2019 Springer Nature Switzerland AG
About this paper
Cite this paper
Peška, L., Trojanová, H. (2019). LineIT: Similarity Search and Recommendation Tool for Photo Lineup Assembling. In: Anderst-Kotsis, G., et al. Database and Expert Systems Applications. DEXA 2019. Communications in Computer and Information Science, vol 1062. Springer, Cham. https://doi.org/10.1007/978-3-030-27684-3_25
Download citation
DOI: https://doi.org/10.1007/978-3-030-27684-3_25
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-27683-6
Online ISBN: 978-3-030-27684-3
eBook Packages: Computer ScienceComputer Science (R0)