{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,3]],"date-time":"2026-02-03T07:43:03Z","timestamp":1770104583171,"version":"3.49.0"},"reference-count":61,"publisher":"Springer Science and Business Media LLC","issue":"2","license":[{"start":{"date-parts":[[2026,2,2]],"date-time":"2026-02-02T00:00:00Z","timestamp":1769990400000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,2,2]],"date-time":"2026-02-02T00:00:00Z","timestamp":1769990400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/501100001652","name":"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg","doi-asserted-by":"crossref","id":[{"id":"10.13039\/501100001652","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Multimed Tools Appl"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Many biometry methods extract task-relevant information from images. In forensic applications, these images may stem from uncontrolled sources like surveillance cameras in the wild. Such devices oftentimes strongly compress the data, which can significantly complicate biometric tasks. This issue is exacerbated by the emergence of AI compression, which may provide visually appealing images that are of questionable value for biometric identification. The purpose of this work is to investigate potential pitfalls of AI compression. We evaluate six AI compression methods including the recently standardized JPEG AI on the four biometric modalities of irises, fingerprints, fabrics and tattoos. We qualitatively show multiple cases when AI compression achieves misleading results. Tattoos in particular includes misrepresentations of color or shapes at strong compression rates. The quantitative evaluation shows impact on recognition rates when there are few identifying features, such as in low-resolution iris images. Further results show that compressors with MSE loss are prone to omit important image details, and MSE+LPIPS loss may hallucinate features. The findings in this paper aim at raising awareness to these pitfalls, and aiding the development robust biometric algorithms for images in the wild.<\/jats:p>","DOI":"10.1007\/s11042-026-21317-3","type":"journal-article","created":{"date-parts":[[2026,2,2]],"date-time":"2026-02-02T18:43:35Z","timestamp":1770057815000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Polished pixels: impact of AI compression on image-based evidence"],"prefix":"10.1007","volume":"85","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-1780-7419","authenticated-orcid":false,"given":"Sandra","family":"Bergmann","sequence":"first","affiliation":[]},{"given":"Denise","family":"Moussa","sequence":"additional","affiliation":[]},{"given":"Christian","family":"Riess","sequence":"additional","affiliation":[]}],"member":"297","published-online":{"date-parts":[[2026,2,2]]},"reference":[{"key":"21317_CR1","volume-title":"Handbook of Biometrics","author":"AK Jain","year":"2007","unstructured":"Jain AK, Flynn P, Ross AA (2007) Handbook of Biometrics. Springer, New York"},{"issue":"2","key":"21317_CR2","doi-asserted-by":"publisher","first-page":"36","DOI":"10.1109\/MC.2010.38","volume":"43","author":"AK Jain","year":"2010","unstructured":"Jain AK, Feng J, Nandakumar K (2010) Fingerprint Matching. Computer 43(2):36\u201344","journal-title":"Computer"},{"issue":"1","key":"21317_CR3","doi-asserted-by":"publisher","first-page":"88","DOI":"10.1109\/TPAMI.2010.59","volume":"33","author":"AK Jain","year":"2011","unstructured":"Jain AK, Feng J (2011) Latent Fingerprint Matching. IEEE Trans Pattern Anal Mach Intell 33(1):88\u2013100","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"issue":"3","key":"21317_CR4","doi-asserted-by":"publisher","first-page":"441","DOI":"10.1109\/TIFS.2015.2480381","volume":"11","author":"A Dantcheva","year":"2016","unstructured":"Dantcheva A, Elia P, Ross A (2016) What Else Does Your Biometric Data Reveal? A Survey on Soft Biometrics. IEEE Trans Inf Forensics Secur 11(3):441\u2013467","journal-title":"IEEE Trans Inf Forensics Secur"},{"key":"21317_CR5","doi-asserted-by":"crossref","unstructured":"Lee J-E, Jain AK, Jin R (2008) Scars, marks and tattoos (SMT): Soft biometric for suspect and victim identification. In: Biometrics symposium","DOI":"10.1117\/2.1200906.1282"},{"key":"21317_CR6","doi-asserted-by":"publisher","first-page":"218","DOI":"10.1016\/j.patrec.2015.08.006","volume":"68","author":"MS Nixon","year":"2015","unstructured":"Nixon MS, Correia PL, Nasrollahi K, Moeslund TB, Hadid A, Tistarelli M (2015) On soft biometrics. Pattern Recogn Lett 68:218\u2013230","journal-title":"Pattern Recogn Lett"},{"key":"21317_CR7","volume-title":"Forensic Handwriting Identification - Fundamental Concepts and Principles","author":"R Morris","year":"2020","unstructured":"Morris R (2020) Forensic Handwriting Identification - Fundamental Concepts and Principles. Elsevier, London, UK"},{"key":"21317_CR8","doi-asserted-by":"crossref","unstructured":"Moussa D, Maier A, Spruck A, Seiler J, Riess C (2022) Forensic license plate recognition with compression-informed transformers. In: 2022 IEEE international conference on image processing (ICIP), pp 406\u2013410","DOI":"10.1109\/ICIP46576.2022.9897178"},{"key":"21317_CR9","doi-asserted-by":"crossref","unstructured":"Ives RW, Bonney BL, Etter DM (2005) Effect of Image Compression on Iris Recognition. In: IEEE instrumentation and measurement technology conference proceedings, vol 3","DOI":"10.1109\/IMTC.2005.1604535"},{"key":"21317_CR10","doi-asserted-by":"crossref","unstructured":"Mascher-Kampfer A, St\u00f6gner H, Uhl A (2007) Comparison of compression algorithms\u2019 impact on fingerprint and face recognition accuracy. In: Visual communications and image processing, vol 6508","DOI":"10.1117\/12.699199"},{"key":"21317_CR11","unstructured":"Matschitsch S, Tschinder M, Uhl A (2007) Comparison of compression algorithms\u2019 impact on iris recognition accuracy. In: Advances in biometrics"},{"key":"21317_CR12","doi-asserted-by":"publisher","first-page":"625","DOI":"10.1007\/978-1-4899-7488-4_49","volume-title":"Encyclopedia of biometrics","author":"M Hara","year":"2015","unstructured":"Hara M (2015) Fingerprint image enhancement. In: Li SZ, Jain AK (eds) Encyclopedia of biometrics. Springer, US, pp 625\u2013634"},{"key":"21317_CR13","doi-asserted-by":"crossref","unstructured":"Toderici G, Vincent D, Johnston N, Jin\u00a0Hwang S, Minnen D, Shor J, Covell M (2017) Full Resolution Image Compression With Recurrent Neural Networks. In: IEEE conference on computer vision and pattern recognition","DOI":"10.1109\/CVPR.2017.577"},{"key":"21317_CR14","unstructured":"Ball\u00e9 J, Minnen D, Singh S, Hwang SJ, Johnston N (2018) Variational image compression with a scale hyperprior. In: International conference on learning representations"},{"key":"21317_CR15","first-page":"11913","volume":"33","author":"F Mentzer","year":"2020","unstructured":"Mentzer F, Toderici GD, Tschannen M, Agustsson E (2020) High-Fidelity Generative Image Compression. Adv Neural Inf Process Syst 33:11913\u201311924","journal-title":"Adv Neural Inf Process Syst"},{"key":"21317_CR16","doi-asserted-by":"crossref","unstructured":"Ascenso J, Akyazi P, Pereira F, Ebrahimi T (2020) Learning-based image coding: early solutions reviewing and subjective quality evaluation. In: SPIE optics, photonics and digital technologies for imaging applications VI, p 113530","DOI":"10.1117\/12.2555368"},{"issue":"4","key":"21317_CR17","doi-asserted-by":"publisher","first-page":"60","DOI":"10.1109\/MMUL.2024.3485255","volume":"31","author":"E Alshina","year":"2024","unstructured":"Alshina E, Ascenso J, Ebrahimi T (2024) JPEG AI: the first international standard for image coding based on an end-to-end learning-based approach. IEEE Multimed 31(4):60\u201369","journal-title":"IEEE Multimed"},{"issue":"1","key":"21317_CR18","doi-asserted-by":"publisher","first-page":"100","DOI":"10.1109\/MMUL.2023.3245919","volume":"30","author":"J Ascenso","year":"2023","unstructured":"Ascenso J, Alshina E, Ebrahimi T (2023) The JPEG AI standard: providing efficient human and machine visual data consumption. IEEE Multimed 30(1):100\u2013111","journal-title":"IEEE Multimed"},{"key":"21317_CR19","doi-asserted-by":"crossref","unstructured":"Hofer N, B\u00f6hme R (2024) A taxonomy of miscompressions: preparing image forensics for neural compression. In: IEEE international workshop on information forensics and security","DOI":"10.1109\/WIFS61860.2024.10810704"},{"key":"21317_CR20","doi-asserted-by":"crossref","unstructured":"Jalilian E, Hofbauer H, Uhl A (2022) Iris image compression using deep convolutional neural networks. Sensors 22(7)","DOI":"10.3390\/s22072698"},{"issue":"11","key":"21317_CR21","doi-asserted-by":"publisher","first-page":"1148","DOI":"10.1109\/34.244676","volume":"15","author":"JG Daugman","year":"1993","unstructured":"Daugman JG (1993) High confidence visual recognition of persons by a test of statistical independence. IEEE Trans Pattern Anal Mach Intell 15(11):1148\u20131161","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"21317_CR22","doi-asserted-by":"crossref","unstructured":"Nguyen K, Proen\u00e7a H, Alonso-Fernandez F (2024) Deep learning for iris recognition: a survey. ACM Comput Surv 56","DOI":"10.1145\/3651306"},{"key":"21317_CR23","doi-asserted-by":"publisher","first-page":"30723","DOI":"10.1109\/ACCESS.2024.3369035","volume":"12","author":"R Gao","year":"2024","unstructured":"Gao R, Bourlai T (2024) On designing a SwinIris transformer based iris recognition system. IEEE Access 12:30723\u201330737","journal-title":"IEEE Access"},{"key":"21317_CR24","unstructured":"Lazarski W, Zieba M, Jeanneau T, Zillig T, Brendel C (2022) Two-headed eye-segmentation approach for biometric identification. arxiv:2209.15471"},{"key":"21317_CR25","doi-asserted-by":"crossref","unstructured":"Yang K, Xu Z, Fei J (2021) DualSANet: Dual Spatial Attention Network for Iris Recognition. In: IEEE winter conference on applications of computer vision, pp 888\u2013896","DOI":"10.1109\/WACV48630.2021.00093"},{"issue":"5","key":"21317_CR26","doi-asserted-by":"publisher","first-page":"877","DOI":"10.1016\/S0031-3203(98)00107-1","volume":"32","author":"A Farina","year":"1999","unstructured":"Farina A, Kov\u00e1cs-Vajna ZM, Leone A (1999) Fingerprint minutiae extraction from skeletonized binary images. Pattern Recognit 32(5):877\u2013889","journal-title":"Pattern Recognit"},{"issue":"1","key":"21317_CR27","doi-asserted-by":"publisher","first-page":"27","DOI":"10.1109\/34.566808","volume":"19","author":"D Maio","year":"1997","unstructured":"Maio D, Maltoni D (1997) Direct gray-scale minutiae detection in fingerprints. IEEE Trans Pattern Anal Mach Intell 19(1):27\u201340","journal-title":"IEEE Trans Pattern Anal Mach Intell"},{"key":"21317_CR28","doi-asserted-by":"publisher","first-page":"47","DOI":"10.1049\/iet-bmt.2019.0017","volume":"9","author":"VH Nguyen","year":"2019","unstructured":"Nguyen VH, Liu J, Nguyen TTN, Kim H (2019) Universal fingerprint minutiae extractor using convolutional neural networks. IET Biometrics 9:47\u201357","journal-title":"IET Biometrics"},{"key":"21317_CR29","doi-asserted-by":"crossref","unstructured":"Han H, Jain AK (2013) Tattoo based identification: Sketch to image matching. In: International conference on biometrics","DOI":"10.1109\/ICB.2013.6613003"},{"key":"21317_CR30","doi-asserted-by":"crossref","unstructured":"Di X, Patel VM (2017) Deep learning for tattoo recognition. Springer, Cham, pp 241\u2013256","DOI":"10.1007\/978-3-319-61657-5_10"},{"key":"21317_CR31","doi-asserted-by":"crossref","unstructured":"Nicolas-Diaz M, Morales-Gonzalez A, Mendez-Vazquez H (2022) Weighted average pooling of deep features for tattoo identification. Multimed Tools Appl 81","DOI":"10.1007\/s11042-022-12516-9"},{"key":"21317_CR32","doi-asserted-by":"crossref","unstructured":"Gonzalez-Soler LJ, Salwowski M, Rathgeb C, Fischer D (2024) TattTRN: template reconstruction network for tattoo retrieval . In: Conference on computer vision and pattern recognition workshops","DOI":"10.1109\/CVPRW63382.2024.00160"},{"key":"21317_CR33","doi-asserted-by":"crossref","unstructured":"Funk W, Arnold M, Busch C, Munde A (2005) Evaluation of image compression algorithms for fingerprint and face recognition systems. In: IEEE SMC information assurance workshop","DOI":"10.1109\/IAW.2005.1495936"},{"key":"21317_CR34","doi-asserted-by":"crossref","unstructured":"Bhowmik D, Elawady M, Nogueira K (2021) Security and forensics exploration of learning-based image coding. In: International conference on visual communications and image processing, pp 1\u20135","DOI":"10.1109\/VCIP53242.2021.9675445"},{"key":"21317_CR35","doi-asserted-by":"crossref","unstructured":"Berthet A, Dugelay J-L (2022) AI-based compression: a new unintended counter attack on JPEG-related image forensic detectors? In: IEEE international conference on image processing, pp 3426\u20133430","DOI":"10.1109\/ICIP46576.2022.9897697"},{"key":"21317_CR36","doi-asserted-by":"crossref","unstructured":"Berthet A, Galdi C, Dugelay J-L (2023) On the impact of AI-based compression on deep learning-based source social network identification. In: International workshop on multimedia signal processing","DOI":"10.1109\/MMSP59012.2023.10337726"},{"key":"21317_CR37","doi-asserted-by":"crossref","unstructured":"Cardenuto JaP, Krinsky J, Nogueira L, Bharati A, Moreira D (2025) Implications of neural compression to scientific images. In: ACM workshop on information hiding and multimedia security, pp 80\u201385","DOI":"10.1145\/3733102.3733148"},{"key":"21317_CR38","unstructured":"Cannas ED, Mandelli S, Popovic N, Alkhateeb A, Gnutti A, Bestagini P, Tubaro S (2024) Is JPEG AI going to change image forensics? arXiv:2412.03261"},{"issue":"12","key":"21317_CR39","doi-asserted-by":"publisher","first-page":"7842","DOI":"10.1109\/TCSVT.2023.3276442","volume":"33","author":"T Chen","year":"2023","unstructured":"Chen T, Ma Z (2023) Toward robust neural image compression: adversarial attack and model finetuning. IEEE Trans Circuits Syst Video Technol 33(12):7842\u20137856","journal-title":"IEEE Trans Circuits Syst Video Technol"},{"key":"21317_CR40","unstructured":"Kovalev E, Bychkov G, Abud K, Gushchin A, Chistyakova A, Lavrushkin S, Vatolin D, Antsiferova A (2024) Exploring adversarial robustness of JPEG AI: methodology, comparison and new methods. arXiv:2411.11795"},{"key":"21317_CR41","doi-asserted-by":"crossref","unstructured":"Bergmann S, Moussa D, Brand F, Kaup A, Riess C (2023) Frequency-domain analysis of traces for the detection of AI-based compression. In: International workshop on biometrics and forensics, pp 1\u20136","DOI":"10.1109\/IWBF57495.2023.10157489"},{"key":"21317_CR42","doi-asserted-by":"publisher","first-page":"41","DOI":"10.1016\/j.patrec.2024.02.015","volume":"180","author":"S Bergmann","year":"2024","unstructured":"Bergmann S, Moussa D, Brand F, Kaup A, Riess C (2024) Forensic analysis of ai-compression traces in spatial and frequency domain. Pattern Recogn Lett 180:41\u201347","journal-title":"Pattern Recogn Lett"},{"key":"21317_CR43","unstructured":"Tsereh D, Mirgaleev M, Molodetskikh I, Kazantsev R, Vatolin D (2024) JPEG AI image compression visual artifacts: detection methods and dataset. arXiv:2411.06810"},{"key":"21317_CR44","doi-asserted-by":"crossref","unstructured":"Mari D, Cavasin S, Milani S, Conti M (2024) Effectiveness of learning-based image codecs on fingerprint storage. In: IEEE international workshop on information forensics and security, pp 1\u20136","DOI":"10.1109\/WIFS61860.2024.10810674"},{"key":"21317_CR45","doi-asserted-by":"crossref","unstructured":"Bousnina N, Ascenso J, Correia PL, Pereira F (2022) Impact of conventional and AI-based image coding on AI-based face recognition performance. In: European workshop on visual information processing, pp 1\u20136","DOI":"10.1109\/EUVIP53989.2022.9922751"},{"key":"21317_CR46","doi-asserted-by":"crossref","unstructured":"Alkhateeb A, Gnutti A, Guerrini F, Leonardi R, Ascenso J, Pereira F (2024) JPEG AI Compressed Domain Face Detection. In: IEEE international workshop on multimedia signal processing, pp 1\u20136","DOI":"10.1109\/MMSP61759.2024.10743984"},{"key":"21317_CR47","unstructured":"Minnen D, Ball\u00e9 J, Toderici GD (2018) Joint Autoregressive and Hierarchical Priors for Learned Image Compression. In: Advances in neural information processing systems, vol 31"},{"key":"21317_CR48","doi-asserted-by":"crossref","unstructured":"Minnen D, Singh S (2020) Channel-wise autoregressive entropy models for learned image compression. In: IEEE international conference on image processing","DOI":"10.1109\/ICIP40778.2020.9190935"},{"key":"21317_CR49","doi-asserted-by":"crossref","unstructured":"Cheng Z, Sun H, Takeuchi M, Katto J (2020) Learned image compression with discretized gaussian mixture likelihoods and attention modules. In: Conference on computer vision and pattern recognition, pp 7936\u20137945","DOI":"10.1109\/CVPR42600.2020.00796"},{"key":"21317_CR50","unstructured":"Yang R, Mandt S (2024) Lossy image compression with conditional diffusion models. In: Conference on neural information processing systems"},{"key":"21317_CR51","doi-asserted-by":"crossref","unstructured":"Duan Z, Lu M, Ma Z, Zhu F (2022) Opening the black box of learned image coders. In: Picture coding symposium","DOI":"10.1109\/PCS56426.2022.10018031"},{"key":"21317_CR52","doi-asserted-by":"crossref","unstructured":"Mohammadi S, Wu Y, Ascenso J (2023) Fidelity-preserving learning-based image compression: loss function and subjective evaluation methodology. In: International conference on visual communications and image processing","DOI":"10.1109\/VCIP59821.2023.10402691"},{"key":"21317_CR53","unstructured":"Blau Y, Michaeli T (2019) Rethinking Lossy Compression: The Rate-Distortion-Perception Tradeoff. In: Proceedings of the 36th international conference on machine learning, vol 97, pp 675\u2013685"},{"key":"21317_CR54","unstructured":"Tan T (2020) CASIA Iris Image Database V4. http:\/\/biometrics.idealtest.org\/"},{"key":"21317_CR55","unstructured":"BioLab (2002) FVC2002 - fingerprint verification competition. http:\/\/bias.csr.unibo.it\/fvc2000\/default.asp"},{"key":"21317_CR56","doi-asserted-by":"crossref","unstructured":"Kampouris C, Zafeiriou S, Ghosh A, Malassiotis S (2016) Fine-grained material classification using micro-geometry and reflectance. In: European conference on computer vision","DOI":"10.1007\/978-3-319-46454-1_47"},{"key":"21317_CR57","unstructured":"Hrka\u0107 T, Brki\u0107 K, Kalafati\u0107 Z (2016) Tattoo detection for soft biometric de-identification based on convolutional neural networks. In: OAGM-ARW joint workshop"},{"key":"21317_CR58","unstructured":"Worldcoin (2023) IRIS: iris recognition inference system of the worldcoin project. https:\/\/github.com\/worldcoin\/open-iris"},{"issue":"1","key":"21317_CR59","doi-asserted-by":"publisher","first-page":"21","DOI":"10.1109\/TCSVT.2003.818350","volume":"14","author":"J Daugman","year":"2004","unstructured":"Daugman J (2004) How iris recognition works. IEEE Trans Circuits Syst Video Technol 14(1):21\u201330","journal-title":"IEEE Trans Circuits Syst Video Technol"},{"key":"21317_CR60","unstructured":"Va\u017ean R (2023) SourceAFIS fingerprint matches. https:\/\/sourceafis.machinezoo.com\/"},{"key":"21317_CR61","doi-asserted-by":"crossref","unstructured":"Yang W, Luo P, Lin L (2014) Clothing co-parsing by joint image segmentation and labeling. In: Conference on computer vision and pattern recognition","DOI":"10.1109\/CVPR.2014.407"}],"container-title":["Multimedia Tools and Applications"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11042-026-21317-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11042-026-21317-3","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11042-026-21317-3.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,2]],"date-time":"2026-02-02T18:43:44Z","timestamp":1770057824000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11042-026-21317-3"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,2]]},"references-count":61,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,2]]}},"alternative-id":["21317"],"URL":"https:\/\/doi.org\/10.1007\/s11042-026-21317-3","relation":{},"ISSN":["1573-7721"],"issn-type":[{"value":"1573-7721","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,2]]},"assertion":[{"value":"23 November 2024","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"29 July 2025","order":2,"name":"revised","label":"Revised","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"12 September 2025","order":3,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"2 February 2026","order":4,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"All authors have no competing interests to declare.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}},{"value":"Not applicable (the work exclusively uses public datasets).","order":3,"name":"Ethics","group":{"name":"EthicsHeading","label":"Ethical approval"}},{"value":"Not applicable.","order":4,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to participate"}},{"value":"Not applicable.","order":5,"name":"Ethics","group":{"name":"EthicsHeading","label":"Consent to publish"}}],"article-number":"95"}}