Finland-based researcher in image/video coding, quality assessment and deep-learning medical imaging; record since 1989
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
72
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-16) 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—
Not yet scored: the output axis needs per-venue publication data not available in this snapshot, so no value is estimated. It would measure career output volume, consistency, and venue distribution.
Impact—
Not yet scored: citation-linked impact records are not connected for this profile, so the axis is left blank rather than estimated. Once available, it would measure citation accumulation, per-paper influence, and field-weighted citation performance.
Focus0/100
The focus score of 0 reflects unusually wide topical coverage rather than deep concentration. OpenAlex topics show five large clusters of comparable size — video coding and compression (241 works), advanced data compression (203), image and video quality assessment (203), image and signal denoising (155), and image retrieval and classification (150) — and the recent field split is an even 0.5/0.5 between machine-vision image coding and deep-learning medical image computing.
Collaboration46/100
The collaboration score of 46 sits mid-range: the snapshot lists 9 distinct coauthors against 125 works in the recent three-year window, pointing to small, stable teams rather than large rotating consortia. Recurring partners include Serkan Kiranyaz, Ozer Can Devecioglu, Farhad Pakdaman, Hamed Rezazadegan Tavakoli, Khoa Pham-Dinh, and Sanaz Nami; on the two reviewed 2026 papers he holds middle and last author positions.
Activity100/100
Activity scores 100, the top of the scale. The counts-by-year series runs uninterrupted from 1989 (2 works) to 2026 (38 works), and the recent three-year window alone contains 125 works. Output stays high late in the series — 48 works in 2024, 39 in 2025 and 38 in 2026 — with earlier peaks of 58 in 2016 and 2021.
Reproducibility55/100
The reproducibility score of 55 is the mean over 2 Evolve-reviewed papers (reproMean: 55, reproPapers: 2), both rated medium repro-readiness. One reports -29.8% BD-rate on Cityscapes but relies on an oracle lambda choice and a single dataset; the other claims up to 14% PSNR gain yet shows low absolute fidelity, one baseline, and regressing axial SSIM.
Research summary
Moncef Gabbouj is a Finland-based researcher (listed affiliation: Nokia) whose OpenAlex-linked record spans 1989 to 2026 and shows sustained high-volume output: an h-index of 72, an i10-index of 379, and 125 works in the most recent three-year window, with 38 to 48 works per year from 2024 to 2026. Topic modelling concentrates on video coding and compression (241 works), advanced data compression (203), image and video quality assessment (203), image and signal denoising (155), and image retrieval and classification (150). The recent field distribution splits evenly (0.5/0.5) between computer vision / image coding for machines and deep-learning medical image computing. Evolve has reviewed two 2026 arXiv papers: "You've Seen Enough: Quality-Constrained Image Coding for Machines," where a bilinear-penalty constrained ICM method yields -29.8% BD-rate on Cityscapes against the baseline (with caveats on oracle lambda choice, single dataset, and no significance statistics), and "Multi-Pass, Multi-View Blended Learning for High-Fidelity Volumetric CT Synthesis from Chest X-Rays," which claims up to 14% PSNR gain for CXR-to-CT synthesis although fidelity stays low (at most 15.5 dB PSNR), with one baseline and regressing axial SSIM. Both papers are rated medium repro-readiness. Across them he appears as a middle and a last author, consistent with a senior contributor in small, stable teams. This profile is name-resolved from public bibliographic data; the output and impact radar axes are not yet scored.
Collaboration pattern
The collaboration score of 46 pairs with a compact recent network: 125 works over the last three years against only 9 distinct coauthors in the snapshot, suggesting small, stable teams rather than wide rotating consortia. Recurring partners include Serkan Kiranyaz, Ozer Can Devecioglu, Farhad Pakdaman, Hamed Rezazadegan Tavakoli, Khoa Pham-Dinh, Sanaz Nami, Rashid Mazhar, Tahir Hamid, and Muhammad Chowdhury. On the two reviewed 2026 papers he sits in middle and last author positions, consistent with a senior supervisory role.
Research keywords
Highlights
Quality-constrained image coding for machines cuts BD-rate by 29.8% on Cityscapes
CXR-to-CT volumetric synthesis claims 14% PSNR gain but fidelity stays low
Timeline
- 1989Earliest works in the OpenAlex counts-by-year series (2 publications), the start of a record that continues unbroken through 2026.
- 2002Annual output jumps to 40 works, the highest count of the 1989-2012 stretch of the series.
- 2013Output reaches 47 works, surpassed only later by the 58-work peaks of 2016 and 2021.
- 2016The series hits 58 works, tied for its highest annual count (matched again in 2021).
- 2026Two arXiv preprints enter the Evolve catalog — quality-constrained image coding for machines and CXR-to-CT volumetric synthesis — both rated medium repro-readiness.
Papers in the Evolve catalog
Frequent co-authors
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 in this snapshot come from OpenAlex (identity: openalex); the radar's output and impact axes are not yet scored, even though an h-index of 72 and i10-index of 379 appear in the raw snapshot.
- The reproducibility score (55) and the paper highlights rest on only two Evolve-reviewed 2026 arXiv preprints, both rated medium repro-readiness; verdicts are Evolve's assessments, not peer review.
- This profile is name-resolved from public bibliographic data linked to ORCID 0000-0002-9788-2323; the listed affiliation is Nokia (Finland).
- Counts by year extend to 2026 and may include partial-year data; topic labels are OpenAlex-assigned and may not match the researcher's own framing.