D-CRISP: explaining object detectors by combining randomized and segment-based perturbations

Vista sencilla de ítem
dc.contributor.authorAndrés Fernández, Alain
dc.contributor.authorSer Lorente, Javier del
dc.date.accessioned2026-01-12T10:44:52Z
dc.date.available2026-01-12T10:44:52Z
dc.date.issued2025-10-21
dc.date.updated2026-01-12T10:44:52Z
dc.descriptionPonencia presentada en la 28th European Conference on Artificial Intelligence, ECAI, celebrada en Boloña, entre el 25 y el 30 de octubre de 2025es
dc.description.abstractExplaining the decisions issued by Machine Learning models for object detection tasks is essential in high-stakes decision making scenarios, such as medical image processing and vehicular perception for autonomous driving. Despite the proliferation of post-hoc perturbation-based methods for generating visual explanations, most eXplainable AI (XAI) approaches rely exclusively on either random image masking or selective segmentation-based occlusion, missing the opportunity to synergistically leverage both strategies in a complementary fashion. In this paper we address this gap by proposing D-CRISP (Detector-Combining Randomized Input and Segment Perturbations), a novel post-hoc explanation method for object detection models. D-CRISP unifies both random and region-based occlusions derived from image segmentation, producing multiscale saliency maps that capture both granular (pixel-level) and semantic (region-level) cues about the objects detected by the model. Experiments on the MS-COCO dataset show that D-CRISP significantly outperforms random-masking approaches in terms of explanation faithfulness and localization, while requiring slightly more computation effort than these methods. At the same time, it achieves comparable or better performance than segmentation-based methods, yet with substantially lower mask generation latencies. These results position D-CRISP as a highly effective and efficient XAI alternative for object detection models, particularly suited for time-constrained applications requiring timely, accurate, and interpretable decisions.en
dc.description.sponsorshipThe authors acknowledge funding support from FaRADAI project (ref. 101103386) funded by the European Commission under the European Defence Fund (EDF-2021-DIGIT-R). Their work is also supported by the Basque Government through the consolidated research group MATHMODE (ref. IT1456-22)en
dc.identifier.citationAndres, A., & Del Ser, J. (2025). D-CRISP: explaining object detectors by combining randomized and segment-based perturbations. Frontiers in Artificial Intelligence and Applications, 413, 217-224. https://doi.org/10.3233/FAIA250809
dc.identifier.doi10.3233/FAIA250809
dc.identifier.eissn1879-8314
dc.identifier.isbn9781643686318
dc.identifier.issn0922-6389
dc.identifier.urihttps://hdl.handle.net/20.500.14454/4680
dc.language.isoeng
dc.publisherIOS Press BV
dc.rights© 2025 The Authors
dc.titleD-CRISP: explaining object detectors by combining randomized and segment-based perturbationsen
dc.typeconference paper
dcterms.accessRightsopen access
oaire.citation.endPage224
oaire.citation.startPage217
oaire.citation.titleFrontiers in Artificial Intelligence and Applications
oaire.citation.volume413
oaire.licenseConditionhttps://creativecommons.org/licenses/by-nc/4.0/
oaire.versionVoR
Archivos
Bloque original
Mostrando 1 - 1 de 1
Miniatura
Nombre:
andres_dcrisp_2025.pdf
Tamaño:
2.82 MB
Formato:
Adobe Portable Document Format
Descargar
Colecciones
Ponencias