University of Florida researcher bridging image analysis, 3D shape work, and ML-based scientific data compression.
2
papers here
1
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
43
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-17) 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—
Output volume is not yet scored in this profile: no computed works-count dimension is available. The axis would measure total indexed publications, recent-year output, and authorship-position spread; the current snapshot simply lacks a value, so no score or estimate is offered.
Impact—
Impact is not yet scored in this snapshot. The axis would measure citation accumulation, citing-work volume, and cross-field influence over time; no computed value exists in the provided data, so this profile deliberately offers no estimate for influence.
Focus100/100
Indexed topics form five tight clusters: medical image segmentation techniques lead with 97 works, followed by traffic control and management (74), 3D shape modeling and analysis (55), traffic prediction and management techniques (51), and image and signal denoising methods (47) — a concentrated vision-and-signal portfolio supporting a focus score of 100.
Collaboration28/100
The recent three-year window records 61 works but only 3 distinct coauthors, yielding a collaboration score of 28. The recurring partners are Liangji Zhu, Sanjay Ranka, and Surya Majumder, indicating a compact, stable core team rather than a broad collaborative network in this period.
Activity100/100
Activity is maximal: countsByYear rises from 23 works in 2020 to 27 in 2023 and 34 in 2024, with 14 works indexed in 2025 and 13 already in 2026, and 61 works across the recent three years — a sustained late-record cadence grounding an activity score of 100.
Reproducibility55/100
Reproducibility rests on a single reviewed paper (reproMean 55, reproPapers 1): the RVQ + pixel-U-Net + GAE pipeline reports NRMSE cuts at equal compression ratio on 3 datasets, but results are figure-only, classical baselines are absent, and the error-bound guarantee is thin; readiness is rated medium.
Research summary
Anand Rangarajan is a researcher at the University of Florida (US), profiled here from public OpenAlex bibliographic data. The snapshot records an h-index of 43 and an i10-index of 137. Indexed topic mass concentrates in medical image segmentation techniques (97 works), traffic control and management (74), 3D shape modeling and analysis (55), traffic prediction and management techniques (51), and image and signal denoising methods (47) — a vision- and signal-heavy portfolio now extending into machine learning for scientific computing. Output stays high late in the record: 61 works in the last three years, with annual counts rising from 23 in 2020 to 27 in 2023 and 34 in 2024, and 13 works already indexed for 2026. The Evolve catalog carries two of his 2026 arXiv papers on scientific data compression. The reviewed one, on which he is last author, combines residual vector quantization, pixel-space U-Net residual correction, and a Guaranteed Autoencoder, reporting lower NRMSE at equal compression ratio across three datasets; Evolve's verdict calls the results promising but notes figure-only evidence, missing classical baselines, and a thin error-bound guarantee, with medium reproducibility readiness. His recent collaboration footprint is compact — 3 distinct coauthors in the window, with Liangji Zhu, Sanjay Ranka, and Surya Majumder the recurring partners.
Collaboration pattern
The recent window is a tight three-person circle: 61 works with only 3 distinct coauthors. Liangji Zhu, Sanjay Ranka, and Surya Majumder recur as frequent coauthors, and both 2026 compression papers continue the pattern — Rangarajan is last author on the reviewed one and corresponding author on neither. The collaboration dimension scores 28, reflecting a compact core team rather than a broad network; the snapshot carries no historical coauthor totals.
Research keywords
Highlights
Error-bounded neural compression pipeline (2026)
Cross-variable latent decorrelation for multivariate data
A second 2026 arXiv paper on multivariate scientific data compression with learned cross-variable latent decorrelation and autoregressive entropy modeling; it is listed in the Evolve catalog without a review verdict, keywords, or reproducibility reading yet.
Late-career output surge
OpenAlex countsByYear climbs from 23 works in 2020 to 27 in 2023 and 34 in 2024 — the highest single-year count in the provided window — with 61 works across the last three years, surpassing the earlier 2005 peak of 23.
Timeline
- 199611 works indexed in the OpenAlex record, a clear early step up from the single-digit annual counts of the surrounding years.
- 200218 works indexed, tripling the 6 works recorded for 2001.
- 200523 works indexed — the highest annual count of the record's first two decades.
- 201614 works indexed, continuing a steady mid-2010s cadence (13 works in 2014).
- 202434 works indexed, the highest single-year count anywhere in the provided countsByYear window.
- 2026Two arXiv papers on ML-based scientific data compression enter the Evolve catalog; Rangarajan is last author on the reviewed one.
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 are drawn from a single OpenAlex snapshot; the h-index (43) and i10-index (137) may lag current citation counts.
- Output and impact dimension scores are not computed in this snapshot, so no estimates are offered for those axes.
- countsByYear includes a lone 1961 entry (1 work) that predates all other indexed years and likely reflects an indexing artifact; treat early-year counts cautiously.
- Only one of the two catalog papers carries an Evolve review verdict and a reproducibility reading (reproPapers: 1); the second is listed without assessment.