About Computational Visual Media

Computational Visual Media

Computational Visual Media (CVM) is an English-language, peer-reviewed open-access journal in computer science, based in Beijing, China. It was founded in 2015 by the Department of Computer Science of Tsinghua University and is published by Tsinghua University Press, for many years in partnership with Springer. Edited by Shi-Min Hu of Tsinghua University, it publishes original research papers and significant review articles on novel ideas, methods and systems for visual media — the field that brings together computer graphics, computer vision, image and video processing, geometric computing and machine learning.

The journal arrived at a moment when visual computing was changing rapidly. Images and video had become the dominant forms of digital content, produced by everything from cameras and phones to medical scanners, satellites and games, and machine learning was beginning to transform how such content is analysed and created. Computational Visual Media was created as a dedicated international venue for that converging field, published from one of China's strongest computer science departments and made freely available to readers everywhere.

Overview and identifying details

The journal's print ISSN is 2096-0433 and its electronic ISSN is 2096-0662. Its standard abbreviation is Comput. Vis. Media, and its articles published with Springer carry DOIs in the 10.1007/s41095 series. Its current journal home is on SciOpen, the publishing platform of Tsinghua University Press, at the journal page for ISSN 2096-0433, and its articles are also available on IEEE Xplore. Bibliographic listings record Tsinghua University Press as publisher, China as the country of publication, and computer science as its subject field.

Computational Visual Media is a single-blind peer-reviewed journal. It is fully open access: the open-access fees, or article-processing charges, are fully sponsored by Tsinghua University, so authors can publish in the journal without paying an article-processing charge themselves. That sponsorship model is one of its defining features — it removes the fee barrier that open-access publishing often places on authors, while keeping every published article free for any reader to access.

Founding in 2015 at Tsinghua University

Computational Visual Media is an English academic journal founded by the Department of Computer Science of Tsinghua University in 2015 and co-published by Tsinghua University Press, under Tsinghua Holdings, and Springer. Its first issue appeared in March 2015 as volume 1, number 1, opening with an editorial message from the editor-in-chief, Shi-Min Hu of Tsinghua's Department of Computer Science and Technology.

That first issue, with eight papers, was a special issue devoted to the 3rd International Conference on Computational Visual Media, held at Tsinghua University in April 2015. The journal and the conference therefore grew up together: the conference series had begun earlier, with the first International Conference on Computational Visual Media held in Beijing in November 2012, its proceedings — 33 revised full papers selected from 81 submissions, edited by Shi-Min Hu and Ralph Martin — published by Springer. By 2015 the community around the conference was ready for a permanent journal, and Tsinghua provided it.

Editorial leadership

The editor-in-chief is Shi-Min Hu, a professor in the Department of Computer Science and Technology at Tsinghua University in Beijing. Hu received his PhD from Zhejiang University in 1996, and his research interests include digital geometry processing, video processing, rendering, computer animation and computer-aided geometric design. He also serves on the editorial boards of journals including Computer Aided Design, Computers & Graphics and the Journal of Computer Science and Technology, placing the journal's leadership inside the wider graphics and geometric-modelling community it serves.

At the journal's launch, its two associate editors-in-chief were Ming C. Lin of the University of Maryland and Ralph R. Martin of Cardiff University, both named in Hu's founding editorial and in Tsinghua's own reporting. That founding team — a Tsinghua editor-in-chief with senior associate editors in the United States and the United Kingdom — signalled from the start that Computational Visual Media was intended as an international journal in authorship, editorship and readership, not a house journal for one department, even though Tsinghua founded and sponsors it.

Scope: graphics, vision and machine learning for visual media

The journal publishes articles that focus on, but are not limited to, a defined set of areas. Its current SciOpen scope lists editing and composition of visual media, geometric computing for images and video, machine learning for visual media, physically based animation, realistic rendering, recognition and understanding of visual media, visual computing for robotics, and visualisation and visual analytics. Its earlier Springer scope was expressed in a longer list that also included 3D visual media processing, classification and composition of visual media, cognition of visual media, content security for visual media, enhancement and re-rendering of visual media, interactive editing, social media computing, understanding of visual media and visual media retrieval.

Underlying both lists is the same idea: other interdisciplinary research into visual media that combines aspects of computer graphics, computer vision, image and video processing, geometric computing and machine learning is also within the journal's scope. In practice, bibliographic analyses of its contents find its work concentrated in computer graphics, artificial intelligence, computer vision and pattern recognition, with individual papers addressing topics such as 3D reconstruction, video tracking, salient object detection and visual analytics for machine learning. Its highly cited content includes survey work — for example, a widely cited survey of salient object detection and a survey of visual analytics techniques for machine learning — reflecting the value its review articles have had for researchers entering fast-moving topics.

The conference and community behind the journal

Computational Visual Media cannot be fully understood apart from the community that shares its name. The International Conference on Computational Visual Media began in Beijing in 2012, and the Tsinghua-based Computational Visual Media Forum, operated by the Computer Graphics Group at Tsinghua University, exists to exchange novel research ideas and significant practical results both underpinning and applying visual media. The forum promotes cross-disciplinary research amalgamating computer graphics, computer vision, machine learning, image processing, video processing, visualisation and geometric computing, and hosts conferences, workshops and public seminars by leading and emerging researchers.

The journal gave that community a permanent publication channel. Its first issue drew directly on the third CVM conference, and conference-journal links of this kind are a familiar pattern in computer graphics and vision, where conferences are the field's primary venues. A journal adds what conferences cannot easily provide: longer, archival versions of work, substantial review articles, and a citable record that is indexed in the major bibliographic databases and counts in formal research evaluation.

Publishing model and frequency

At its launch the journal was published quarterly, and its Springer-era volumes follow that pattern clearly: volume 1 in 2015 ran from March to December in four issues, and the sequence continued to volume 10 in 2024. Its present SciOpen listing describes it as an open-access journal published bimonthly by Tsinghua University Press, with all articles available on IEEE Xplore — a faster cadence that reflects its growth. Tsinghua University Press announced in connection with the journal's current platform that it reports a first-decision time of 28 days and publication in 60 days; such figures are the publisher's current service indicators rather than guarantees for individual manuscripts.

The publishing partnership has also evolved. For its first decade the journal was published by Tsinghua University Press with Springer, under Springer's SpringerOpen open-access programme, and its back volumes remain searchable through Springer's site, which now notes that the journal is archived there and no longer receiving submissions with that publisher. Its live home is SciOpen, Tsinghua University Press's own platform, at the website recorded in its directory listing. The continuity through that transition is the title, the ISSN and the Tsinghua editorship; the platform changed, the journal did not.

Indexing and recognition

Computational Visual Media was included in the Science Citation Index Expanded database of Web of Science on 29 April 2021, as Tsinghua University Press's parent group reported at the time. In the six years since its founding, the journal had by then been included in 24 well-known databases at home and abroad, including SCI, EI (Engineering Index), DBLP, INSPEC, Scopus, DOAJ and OCLC. Its directory description likewise records indexing in SCI-Expanded, EI and Scopus.

The journal also received early institutional backing within China. At its establishment it received funding from the China Science and Technology Journals' International Influence Improvement Program, and in 2019 it was selected into the China Science and Technology Journal Excellence Action Plan hosted by the China Association for Science and Technology. Its current publisher listings report strong recent citation indicators for a journal founded only in 2015. Those markers together explain its trajectory: a university-sponsored open-access title that moved within six years from launch to Science Citation Index Expanded coverage.

Who publishes in it and who reads it

The journal's natural authors are researchers in computer graphics, computer vision, image and video processing, geometric modelling and processing, visualisation and machine learning — in universities, research institutes and technology companies. Bibliographic profiles of the journal show Tsinghua University as its most frequent contributing affiliation, which is expected for a journal founded there, but its authorship and editorial board are international, and its founding associate editors were based at Cardiff University and the University of Maryland.

Its readers overlap with its authors, as in most research journals, but its open-access, no-author-fee model widens that readership considerably. A practitioner, student or researcher at an institution without a subscription budget can read everything it publishes, and an author without grant funding for article-processing charges can still submit. In a field where much of the literature sits either behind paywalls or behind author fees, that combination is a genuine part of the journal's appeal, and it is made possible by Tsinghua University's sponsorship of its publication costs.

Place among visual computing journals

Visual computing already had established journals when Computational Visual Media launched in 2015, including long-running graphics and vision titles from commercial and society publishers. What Tsinghua offered was different in structure: a journal owned and sponsored by a university, edited from a leading computer science department, free to read and free to publish in, and explicitly scoped across the whole of visual media rather than only graphics or only vision. Its scope statement's insistence on interdisciplinary work — combining graphics, vision, image and video processing, geometric computing and machine learning — anticipated how thoroughly those fields have since merged, particularly as machine learning methods became central to both image understanding and image creation.

That positioning also reflects a broader shift in Chinese academic publishing, in which leading universities have built English-language, internationally indexed journals in their strongest fields rather than relying only on overseas titles. Computational Visual Media, founded the same decade as many such titles, is the visual-computing example: Beijing-based, Tsinghua-edited, and aimed at a global research community from its first issue.

Why the journal matters

Computational Visual Media matters because it gave a fast-converging field a journal shaped around the convergence itself. Since March 2015 it has published original research and significant reviews on editing and composition of visual media, geometric computing, machine learning for visual media, physically based animation, realistic rendering, recognition and understanding of visual media, visual computing for robotics, and visualisation and visual analytics — the working agenda of modern visual computing — under editor-in-chief Shi-Min Hu of Tsinghua University.

Its identifying facts are compact: founded in 2015 by Tsinghua University's Department of Computer Science in Beijing, published by Tsinghua University Press, print ISSN 2096-0433 and electronic ISSN 2096-0662, open access with publication costs sponsored by Tsinghua University, and indexed in SCI-Expanded, EI and Scopus. For researchers, librarians and readers mapping China's contribution to computer science publishing, it stands as one of the clearest examples of a Chinese university journal achieving international standing within a few years of its launch — built not on a legacy title inherited from the past, but on a research community, a conference series and a sponsor willing to pay for open access.