University of Klagenfurt researcher building generative-AI and semantic streaming methods for adaptive video delivery
2
papers here
2
reviewed by Evolve
0
first-author
0
corresponding
—
works · OpenAlex
16
h-index · OpenAlex
Bibliometric figures come from OpenAlex (2026-10-14) 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—
This axis is unscored: the snapshot lacks a verified total-output record, such as a full works list with venue types. It would measure cumulative publication volume per year and how the researcher's output has grown across the career.
Impact—
Not yet scored: while an h-index of 16 and an i10-index of 31 are on record, the citation aggregates a robust impact measure needs — total citations and field-normalized percentiles — are absent from this snapshot.
Focus0/100
Scores 0/100, signalling breadth rather than a narrow specialty. The field distribution splits evenly (50/50) between multimedia computing (video streaming and generative computer vision) and multimedia systems / video coding, and the five largest topic clusters — image and video quality assessment (45), network traffic and congestion control (32), caching and content delivery (30), energy-efficient wireless sensor networks (26), and mobile ad hoc networks (18) — span streaming, networking, and sensing.
Collaboration32/100
Scores 32/100 on this radar. OpenAlex lists only 4 distinct coauthors in the recent window, and the recurring partners form a tight Klagenfurt group: Emanuele Artioli, Christian Timmerer, Daniele Lorenzi, and Shivi Vats. Both 2026 catalog papers carry middle, non-corresponding authorship within that team, so the collaboration footprint is deep but narrow — concentrated in one working group rather than dispersed across institutions.
Activity100/100
Scores 100/100, the profile's strongest axis. OpenAlex counts 33 works over the last three years, and yearly totals climb from 1–5 in 2012–2019 to 9 (2021), 12 (2022), 8 (2023), 11 (2024), 12 (2025), and 10 already in 2026 — sustained, near-peak productivity across the past six years.
Reproducibility60/100
Scores 60/100, averaged from 2 reviewed papers (reproMean 60). The split is stark: PRESLEY earns high reproducibility readiness with rigorous statistics, BD-rate comparison against ELVIS, and honest limitations, while GenStream is rated low because it is only partially built and its >99.9% bandwidth-saving claim is unquantified. Reproducibility therefore depends heavily on which line of work is examined.
Research summary
Farzad Tashtarian is a researcher at the University of Klagenfurt (AT) working across two evenly split fields — multimedia computing (video streaming and generative computer vision) and multimedia systems / video coding. The OpenAlex snapshot (ORCID 0000-0002-5584-6690) records an h-index of 16 and an i10-index of 31, with 33 works over the last three years and annual counts rising from single digits before 2020 to 9–12 per year since 2021, the basis of a perfect 100/100 activity score. Topic clusters span image and video quality assessment (45), network traffic and congestion control (32), caching and content delivery (30), energy-efficient wireless sensor networks (26), and mobile ad hoc networks (18), while the focus score of 0/100 reflects this breadth. Recent work pivots to generative-AI streaming: the 2026 preprint GenStream streams pose and viewpoint metadata instead of video, claiming >99.9% bandwidth savings versus HEVC, and PRESLEY restores deliberately degraded blocks with conditioned super-resolution, beating ELVIS at matched rate with high reproducibility readiness. Collaboration is concentrated — 4 distinct recent coauthors (Artioli, Timmerer, Lorenzi, Vats), collaboration score 32/100 — with middle, non-corresponding authorship on both reviewed papers. Output and impact axes remain unscored for lack of data.
Collaboration pattern
Collaboration is concentrated: OpenAlex shows 4 distinct recent coauthors, and the recurring names — Emanuele Artioli, Christian Timmerer, Daniele Lorenzi, Shivi Vats — form a single Klagenfurt multimedia-streaming group. On both reviewed 2026 papers Tashtarian appears as a middle author without corresponding role, consistent with a contributing specialist inside a stable team rather than an independent pivot; collaboration radar score: 32/100.
Research keywords
Highlights
PRESLEY: generative refinement beats ELVIS at matched rate
GenStream: >99.9% bandwidth savings via semantic streaming
Output surge: 33 works in three years, h-index 16
OpenAlex counts 33 works over the last three years and yearly totals of 9–12 since 2021 (12 in 2025, 10 already in 2026), driving the perfect activity score alongside an h-index of 16 and an i10-index of 31.
Timeline
- 2026GenStream preprint posted to arXiv, proposing semantic streaming of pose and viewpoint metadata for human-centric media with claimed >99.9% bandwidth savings versus HEVC.
- 2026PRESLEY preprint posted to arXiv, showing conditioned super-resolution refinement beating ELVIS at matched rate, with high reproducibility readiness per Evolve.
- 2026OpenAlex indexes 33 works in the trailing three years and 10 works in 2026, extending the peak run of 9–12 annual works that began in 2021.
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.
- Bibliometric indicators (h-index 16, i10-index 31, yearly work counts, topics) come from an OpenAlex snapshot; identity is name-resolved from public bibliographic data.
- Only 2 papers have been reviewed by Evolve (both 2026 arXiv preprints with middle, non-corresponding authorship), so the output and impact axes are unscored for lack of data.
- Both reviewed works are preprints; peer-review status and final venues are not captured in this snapshot.
- The reproducibility score (60) averages only the 2 reviewed papers and may not represent other outputs.