Przemys{\l}aw Spurek

—

Unclaimed · AI-structured profileOpenAlex

Computer-vision researcher in explainable AI: post-hoc segmentation explanations and novel view synthesis.

papers here

reviewed by Evolve

first-author

corresponding

—

works · OpenAlex

h-index · OpenAlex

Bibliometric figures come from OpenAlex (2026-09-20) and cover the author's entire public record, not just the papers listed here. Reproducibility signal comes from Evolve's own reviews.

Research profile

Output—

Insufficient data: cumulative publication volume, career output trajectory, and a reliable h-index (recorded as a 0 placeholder) are not available in this snapshot. This axis would measure total and per-year publication counts across the researcher's full career, so no score is given.

Impact—

Insufficient data: citation counts, i10-index, and citing-work statistics are unavailable, so scholarly influence cannot be scored. This axis would measure total citations, citation trajectory, and uptake of the two 2026 preprints by other researchers once such metrics are collected.

Focus100/100

The focus dimension scores 100: OpenAlex classifies the single classified work entirely within computer vision (explainable AI / interpretable deep learning), giving a maximally concentrated field distribution. The topic profile — advanced neural network applications, graphics and visualization, advanced vision and imaging, GANs and image synthesis, and multimodal machine learning, one work each — reinforces a coherent interpretability-plus-visual-synthesis agenda across the two 2026 preprints.

Collaboration39/100

The collaboration dimension scores 39. OpenAlex records 6 distinct coauthors, and both 2026 preprints place Spurek in the last author position without corresponding-author status, a seat often tied to senior supervision. Recurring collaborators include Miłosz Adamczyk, Tymoteusz Zapala, Piotr Borycki, Weronika Jakubowska, and Maciej Zięba, suggesting a compact, stable team; the modest score reflects a small network rather than broad multi-group collaboration.

Activity32/100

The activity dimension scores 32, reflecting low recent volume: countsByYear shows 2 works, both in 2026, and recent3yWorks is 2, both arXiv preprints with Spurek as last author. The snapshot captures no earlier years, so it is unclear whether this is an emerging burst of output or a partial record; the score should be read against that thin coverage.

Reproducibility55/100

The reproducibility dimension scores 55 (reproMean), computed from just 1 assessed paper. PiPS carries a medium reproducibility readiness: Evolve's review credits a sound design that exactly preserves mIoU while flagging thin key baselines and faithfulness evaluation. With only one reviewed work, this is an early, narrow signal rather than a stable estimate of overall reproducibility practice.

Research summary

Przemysław Spurek is a computer-vision researcher whose current output centers on explainable AI and scene-level vision. This profile is name-resolved from OpenAlex bibliographic data; the snapshot records no institution or country. The recent record is small but coherent: two 2026 arXiv preprints, both with Spurek in the last author position. PiPS proposes fully post-hoc prototype explanations for interpretable semantic segmentation with exactly preserved mIoU; Evolve's review judges the design sound while flagging thin key baselines and faithfulness evaluation, with medium reproducibility readiness. ORCA extends the agenda to occlusion-aware refinement and completion for novel view synthesis but has not yet been reviewed. OpenAlex field distribution places the single classified work in computer vision (explainable AI / interpretable deep learning), yielding a focus score of 100, and the topic profile spans advanced neural network applications, graphics and visualization, vision and imaging, GANs and image synthesis, and multimodal machine learning, one work each. Collaboration (39) and activity (32) dimensions are modest: 6 distinct coauthors and 2 works in the last three years. Output and impact axes are marked missing because h-index and citation metrics are unavailable, so career seniority and influence cannot yet be assessed from this snapshot.

Collaboration pattern

Both catalog papers list Spurek in the last author position (non-corresponding), a seat typically associated with senior or supervisory roles. OpenAlex records 6 distinct coauthors; recurring collaborators include Miłosz Adamczyk, Tymoteusz Zapala, Piotr Borycki, Weronika Jakubowska, and Maciej Zięba. The collaboration dimension scores 39, consistent with a compact, stable team rather than a broad multi-group network. All observed collaboration comes from the two 2026 preprints; no cross-institution data is in the snapshot.

Research keywords

explainable AIsemantic segmentationprototype-based explanationspost-hoc interpretabilityinterpretable deep learningorthogonal transformationDeepLabV3occlusion-aware refinementimage completionnovel view synthesisgenerative adversarial networksmultimodal machine learning

Highlights

PiPS: first fully post-hoc prototype explanations for segmentation
Evolve's review calls PiPS the first fully post-hoc prototype explanation method for semantic segmentation, with mIoU exactly preserved; the design is judged sound, though key baselines and faithfulness evaluation remain thin, and reproducibility readiness is rated medium.
ORCA: occlusion-aware refinement and completion for novel view synthesis
The second 2026 arXiv preprint, again with Spurek in the last author position, extends the agenda from interpretability to occlusion-aware novel view synthesis and completion; it is listed in the Evolve catalog but has not yet been reviewed.
A coherent 2026 agenda from a senior author seat
Both catalog papers are 2026 arXiv preprints listing Spurek last among the authors, alongside 6 distinct coauthors — a compact pattern consistent with a supervisory role across two directions: interpretable segmentation and view synthesis.

Timeline

2026PiPS: Post-Hoc Prototypical Explanations for Interpretable Semantic Segmentation posted to arXiv, with Spurek in the last author position.
2026ORCA: Occlusion-Aware Refinement and Completion for Novel View Synthesis posted to arXiv, again listing Spurek as the last author.
2026OpenAlex records 2 works in 2026 with 6 distinct coauthors; the sole classified work falls in computer vision (explainable AI / interpretable deep learning).

Papers in the Evolve catalog

Frequent co-authors

Mi{\l}osz Adamczyk×1 Tymoteusz Zapala×1 Piotr Borycki×1 Weronika Jakubowska×1 Maciej Zi\k{e}ba×1 Przemys{\l}aw Spurek×1

About this profile

This profile is AI-structured by Evolve.Pub from public bibliographic data (OpenAlex) and Evolve's own paper reviews. It is not written or endorsed by the researcher unless the profile is claimed. Radar dimensions are computed from real bibliometrics with fixed formulas — raw figures are shown alongside; missing dimensions mean insufficient data, never zero. Errors? Claim the profile to correct or hide it.

Bibliometrics (topics, field distribution, coauthor and recency counts) come from OpenAlex only; h-index and citation-based impact metrics are unavailable, so the output and impact axes are left unscored.
No institution, country, or ORCID is present in the snapshot; the profile is name-resolved from public bibliographic data and may not fully disambiguate the individual.
Only one paper (PiPS) carries an Evolve review; ORCA is cataloged but unreviewed, and the reproducibility mean of 55 rests on a single assessed paper.
The snapshot captures only 2026 arXiv preprints; earlier career history and publication record are outside its scope.