Prompt engineering for synthetic deep-sea mining video: A controlled evaluation of four text-to-video systems
Ocean · 04.11.2026
DOI: 10.26599/ocean.2026.9470033 ↗Kurz gefasst
Fachartikel in der Zeitschrift Ocean, online erschienen am 04.11.2026. Verfasst von Nan Li und 4 weiteren Autor:innen.
Abstract (Originalsprache)
Deep-sea mining video is scarce, and low illumination, sediment plumes, and moving equipment challenge visual monitoring. We evaluated four text-to-video systems on six VBench dimensions, first in an archived 300-video system-level comparison and then in a 240-video extension crossing four prompt conditions, five scenarios, and three requested seeds. Within-system prompt contrasts varied by metric; detailed or negatively constrained prompts did not consistently raise VBench scores. All 15 Wan pairs generated with and without a negative-prompt request had pixel-identical decoded frames, leaving that contrast uninformative for this service. In a separate 60-video review with system identities masked, three raters’ majority labels identified at least one domain violation in 37 clips (61.7%; descriptive Wilson 95% interval, 49.0–72.9%); agreement differed by criterion. These results support VBench scoring and domain-rule review as complementary first-stage screens, while physical fidelity and downstream utility remain to be tested.
Zitieren
APA
Li, N., Yu, L., Huang, C., Yang, Y., & Liu, K. (2026). Prompt engineering for synthetic deep-sea mining video: A controlled evaluation of four text-to-video systems. Ocean. https://doi.org/10.26599/ocean.2026.9470033
Vancouver
Li N, Yu L, Huang C, Yang Y, Liu K. Prompt engineering for synthetic deep-sea mining video: A controlled evaluation of four text-to-video systems. Ocean. 2026. doi:10.26599/ocean.2026.9470033
BibTeX
@article{Li2026,
author = {Li, Nan and Yu, Long and Huang, Chaojun and Yang, Yi and Liu, Kanglin},
title = {Prompt engineering for synthetic deep-sea mining video: A controlled evaluation of four text-to-video systems},
journal = {Ocean},
year = {2026},
doi = {10.26599/ocean.2026.9470033}
}
EUVIMED