Hui Cao- Zhejiang Normal University, Jinhua, China
With the development of computer technology, the digitization of traditional painting has accelerated. Artificial intelligence–generated content (AIGC) and virtual reality (VR) are reshaping the production, dissemination, and aesthetic experience of traditional art, yet a systematic understanding of how they interact with core artistic principles—such as brush–ink logic, spatial conception, and aesthetic intentionality—remains limited. Adopting a systematic and thematic review approach, this paper examines: (1) the technological, aesthetic, and cultural dimensions of digital transformation, showing how digital change extends beyond medium substitution to shifts in viewing logic and meaning construction; (2) how AIGC advances brushstroke simulation, style transfer, and artistic conception but continues to struggle with deeper principles such as spiritual vitality and intention-before-brush; and (3) how VR reconstructs spatial logic and viewing modes in classical painting, revealing tensions between embodied immersion and traditional contemplative, multi-perspective viewing. Drawing on 32 representative studies, this review synthesizes developments in GAN- and diffusion-based generative models as well as VR-based reconstruction and exhibition methods. While AIGC and VR broaden participation by enabling technically trained researchers to engage with traditional culture, the literature also indicates a persistent risk of reproducing surface stylistic features while overlooking essential cultural and spiritual connotations. The paper concludes by identifying key conceptual gaps and proposing future research directions, including mechanism-level modeling of brush-and-ink aesthetics, VR presentation strategies aligned with scattered perspective, and culturally grounded frameworks for human–machine co-creation.
1 Introduction
Chinese traditional art encompasses painting, calligraphy, music, paper-cutting, sketching, and opera, among which occupies a pivotal position. Its aesthetic system is rooted in philosophical thought, cultural identity, and the literati tradition, its connotations extending far beyond visual expression. Classical theory requires painters to view the universe, history, and human life as an organic whole, enabling their brushwork to resonate with the spirit of the times. When artistic creation resonates with a broader knowledge system and socio-cultural context, it can give birth to breathtaking masterpieces. These principles constitute the core criteria for judging artistic creation and are key to understanding the evolution of artistic creation in the contemporary environment.
Although academia has conducted extensive research on painting styles, techniques, historical inheritance, and creative principles, existing research primarily focuses on the works themselves. Current research rarely explores the impact of contemporary technological conditions on,whether in terms of creative subjectivity, aesthetic judgment, dissemination modes, or the cultural connotations inherent in artistic practice. This research gap reflects a tendency: people often simply regardtraditional Chinese painting as a certain style or form, neglecting its philosophical depth and unique aesthetic logic, qualities that distinguish it from other painting traditions.
With the expansion of digital art, driven by computational methods, interactive systems, and immersive environments, traditional Chinese painting faces immense pressure and unprecedented opportunities. Technologies such as AI-generated content (AIGC) and virtual reality (VR) are not merely introducing new tools; they challenge the fundamental aesthetic logic and cultural foundations that have shaped this tradition for centuries. To understand how these technological conditions specifically facilitate this practice, it is necessary to examine how the core attributes of traditional Chinese painting, its brushwork logic, and cultural symbolism interact with emerging digital infrastructure. Therefore, we systematically review the literature on AIGC and virtual reality, organizing previous work according to how these technologies interact with the aesthetic and cultural dimensions of traditional Chinese painting.
To understand these relationships, it is necessary to clarify several fundamental concepts frequently mentioned in this study. “Qi Yun”, often translated as “spiritual resonance” or “life rhythm,” describes the dynamic vitality that imbues the painted world, connecting the artist’s inner state with the vitality of the depicted object. “Yi Jing” refers not only to “artistic conception” but also to a world-building mode in which emotions, ideas, scenery, and symbolism intertwine to create an experiential vision of “image beyond image.” Chinese classical painting theory, especially Xie He’s “Six Principles,” provides the theoretical foundation for elucidating key concepts such as “spirit resonance,” “artistic conception,” and the “core spirit” of painting. Therefore, the so-called “core spirit of traditional painting” is not the core spirit of traditional painting.
While the application of artificial intelligence in this field is rapidly developing, current efforts mainly focus on style simulation or the automated generation of virtual paintings. Deep integration with the framework of traditional culture remains limited. Therefore, AIGC (Artificial Intelligence Generated Content) lacks the ontological foundation of “intention preceding brushwork,” nor does it possess the continuity of “heart-hand-heaven” emphasized by classical aesthetics. This raises an important question: Will the randomness, sampling variability, and data-driven nature of AIGC weaken the intentionality, contemplation, and spiritual responsiveness of traditional Chinese painting?
Similarly, VR introduces a perceptual paradigm with a spatiotemporal logic drastically different from traditional viewing methods. Forms such as handscrolls and landscape paintings rely on dispersed perspectives, the unfolding of time, and contemplative wandering (armchair travel). Viewers experience landscapes not through sensory saturation, but as an imaginative journey. However, virtual reality (VR) organizes visuals through immersive experiences, 360-degree spatial continuity, and high-intensity sensory immersion. These differences raise the following question: Will immersive experiences fundamentally change or even negate the contemplative “wandering” aesthetics at the core of traditional Chinese painting experiences?
To better understand these questions, the “digital transformation” in this study refers not only to the adoption of new media but also to the multi-layered reconstruction of artistic practice, specifically including:
1. Media transformation: from the materiality of handcrafted creation to the virtuality generated by algorithms;
2. Transformation of creative logic: from conscious, figurative creation to probabilistic, data-driven production;
3. Shift in aesthetic system: from contemplative, associative, and symbolic perception to immersive, highly realistic, and realistic sensory environments.
Existing research has explored artificial intelligence-generated content (AIGC) and virtual reality (VR) technologies within a creative context, but most literature still focuses on the technical aspects, lacking in-depth discussions of Chinese aesthetic philosophy, painting theory, or cultural ontology. Meanwhile, few studies explore how the generative randomness of artificial intelligence interacts with traditional intentionality, or how the immersive continuity of virtual reality challenges the temporal layering and fragmented viewing methods inherent in traditional handscrolls.
To address these research gaps, this study aims to explore the following questions:
1. How does AIGC reshape the creative subjectivity inherent in, particularly the principle of “conception preceding creation”?
2. How does VR reconstruct the spatiotemporal and experiential foundations of traditional Chinese painting aesthetics?
3. How do AIGC and VR jointly challenge, reinterpret, or reconstruct the core spirit of,that is, far beyond stylistic characteristics, referring to a complete aesthetic system, including technique, emotion, philosophical cultivation, and cosmology?
Therefore, this paper explores how AIGC and VR influence the creative methods, aesthetic concepts, and value systems of traditional Chinese painting, and whether traditional art can maintain its spirit of brush and ink in digital transformation. Through this exploration, we aim to provide a theoretical foundation and analytical perspective for understanding the transformation of traditional Chinese painting in the digital art era.
2 Review methodology
This study adopts a systematic and thematic review approach to examine how Artificial Intelligence–Generated Content (AIGC) and Virtual Reality (VR) technologies interact with the aesthetic principles, creative logic, and experiential structures of traditional Chinese painting. Rather than conducting empirical experiments or quantitative meta-analysis, the review is designed as a conceptually oriented systematic review, aiming to synthesize interdisciplinary literature across computer science, digital art, human–computer interaction, cultural heritage, and Chinese art theory.
2.1 Literature search strategy
The literature search was conducted across major academic databases commonly used in digital art, computational aesthetics, and cultural heritage research, including Web of Science, Scopus, ACM Digital Library, IEEE Xplore, and selected high-impact journals in art theory and digital humanities. Searches focused on publications from approximately 2015 to 2024, reflecting the period in which deep learning–based image generation and immersive VR technologies became widely applied in artistic and cultural contexts.
Key search terms included combinations of: AIGC, artificial intelligence, generative models, GAN, diffusion model, virtual reality, immersive media, traditional Chinese painting, ink painting, brush-and-ink aesthetics, scattered perspective, digital heritage, and computational aesthetics. Reference lists of relevant articles were also manually reviewed to identify additional influential studies not captured through keyword searches.
2.2 Inclusion and selection criteria
Studies were included if they met one or more of the following criteria: (1) applied AIGC, generative algorithms, or VR technologies to traditional Chinese painting or closely related visual traditions; (2) addressed brushstroke simulation, ink diffusion, style transfer, spatial reconstruction, or immersive viewing in relation to Chinese painting aesthetics; or (3) provided theoretical, cultural, or aesthetic analysis relevant to the interaction between digital technologies and traditional painting systems. Purely technical studies without cultural or artistic relevance, as well as works focusing exclusively on Western painting traditions without transferable aesthetic discussion, were excluded.
Based on these criteria, 32 representative studies were selected for in-depth analysis. The selected corpus balances technical research, applied systems, and theoretical discussions, allowing for a comprehensive view of both technological capabilities and aesthetic implications.
2.3 Analytical framework
The analysis employed thematic synthesis and conceptual mapping rather than statistical aggregation. Selected studies were first grouped according to technological orientation (AIGC vs. VR) and then further categorized by research focus, such as technique-oriented generation, application-oriented systems, genre-specific painting studies, immersive spatial reconstruction, and experiential or cognitive evaluation. These categories were subsequently examined through the lens of core concepts in traditional Chinese painting theory, including qiyu, yijing, yi xian yu bi, and scattered perspective.
Through this framework, the review identifies recurring patterns, conceptual tensions, and methodological gaps between contemporary digital technologies and the philosophical foundations of traditional Chinese painting. The goal is not to rank technological performance, but to clarify how different modes of digital reconstruction reshape creative subjectivity, aesthetic judgment, and viewing experience within a culturally grounded artistic system.
3 AIGC and the generative shift in
Recent developments in AIGC (AI-generated content) have been driven by advances in generative adversarial networks (GANs), diffusion models, and large transformer-based architectures. Early AIGC systems were constrained by LSTM- and RNN-based generative networks (Lee et al., 2020), but the introduction of Transformer architectures and the global diffusion of ChatGPT in 2022 marked a turning point, demonstrating the scalability of pretrained models and accelerating worldwide interest in algorithmic image generation (Grant and Metz, 2022). Within this technological paradigm shift, traditional Chinese painting, especially its highly codified Gongbi tradition,has become a significant subject of inquiry. Alongside research on style transfer, brushstroke simulation, and emotion modeling. Recent advances in diffusion-based architectures have further accelerated this shift, particularly with the release of Stable Diffusion 3, which significantly improves semantic alignment and fine-grained style control in artistic generation (Stability AI, 2024). Moreover, recent studies highlight that diffusion models are increasingly viewed as the dominant paradigm for visual art creation, offering improved controllability, texture fidelity, and sampling stability (Wang B. et al., 2025). Meanwhile,a more fundamental question has emerged: to what extent can computational generation meaningfully engage with the core aesthetic concepts of, such as qiyun (spiritual vitality), yijing (artistic conception), and the classical requirement that “idea precedes brush”?
To bring conceptual clarity to this growing body of work, existing research can be organized into three major categories:
1. Technique-Oriented Approaches: GANs, Transformers, and Diffusion Models.The first type of research focuses on using generative models to replicate style features.GAN-based models have received notable attention. He et al. (2018) introduced ChipGAN, which embeds constraints for brushstrokes, ink diffusion, tonal variation, and compositional gaps. Their results show visually coherent ink diffusion and moderately accurate brushstroke rendering. Sheng et al. (2019) proposed the CPST algorithm, which incorporates four constraints, brushwork, diffused ink, spatial preservation, and yellowing,to enhance the visual authenticity of generated ink paintings. Also recently, Wang X. et al. (2025) developed InkArtGAN, an end-to-end framework that improves detail fidelity, cultural symbol integration, and multi-modal coherence (e.g., poetry, calligraphy, seal carving). While technically impressive, these models still prioritize surface-level stylistic reproduction, and questions remain regarding whether they capture deeper aesthetic principles such as qiyun. Classical Chinese aesthetics articulated in Xie He’s Six Principles, identify qiyun as the core evaluative dimension of painting, referring not only to the vivid appearance of forms but to the dynamic alignment between the artist’s cultivated intention and the spiritual vitality of the depicted world. Current models do not yet engage with this deeper ontological dimension, raising concerns that algorithmic imitation may risk flattening aesthetic meaning even when the visual output appears convincing.Beyond style replication, controllable diffusion models such as ControlNet have introduced spatially aligned conditioning, using edge maps, depth, pose, or line drawings, to guide generation with unprecedented precision (Zhang et al., 2023).As attention continues to grow on Transformer and Diffusion models, researchers are beginning to explore broader generative capabilities beyond GAN architectures. For example, some recent Diffusion-based frameworks, such as Stable Diffusion fine-tuned on ink painting datasets, have demonstrated better global consistency and controllable texture synthesis capabilities, but systematic evaluations of their cultural and philosophical relevance remain limited. Recent progress in controllable diffusion models has further expanded these capacities. ControlNet introduces a mechanism for injecting spatial conditioning, such as line drawings, depth maps, or edge structure, directly into a pretrained diffusion model, thereby enabling fine-grained regulation of composition and brushstroke layout in a way that is directly compatible with Gongbi’s line-driven grammar (Zhang, Rao and Agrawala, 2023). Similarly, the same applies to Transformer; however, these studies have not explicitly assessed whether such models internalize the deeper aesthetic principles of traditional Chinese painting. Therefore, concerns remain that these algorithms may diminish cultural transmission.
2. Application-Oriented Approaches: Style Transfer, Full-Process Simulation, and Emotion Understanding.A second body of work focuses on specific artistic tasks:Style transfer remains the most developed area. Wang X. et al. (2024) argue that due to cultural divergence between Western and Chinese visual systems, conventional style-transfer algorithms fail to preserve core attributes such as “white space,” ink gradients, and rhythmic brush movement. Their method introduces ink-loss constraints and adaptive style modules to better align with traditional Chinese painting aesthetics. Full-process simulation and educational systems aim to emulate brush dynamics and teach painting techniques. Also recent progress further demonstrates that large diffusion models can be adapted for stylistic transformation without any additional training. Style injection techniques propose a training-free mechanism that transfers visual style through manipulating attention layers, enabling culturally specific brush textures or ink tonalities to be integrated even under data-scarce conditions. Recent work on training-free or minimally-trained diffusion style injection offers additional flexibility, allowing traditional Chinese painting styles to be incorporated without full model retraining (Chung et al., 2023). Chu and Tai (2004) developed a virtual 3D brush capable of generating realistic brush footprints by modeling brush geometry, pressure, and motion. Wong and Ip (2000) further modeled stroke formation and ink deposition, enabling accurate reproduction of calligraphic writing styles. These simulation studies contribute to heritage preservation but tend to emphasize physical accuracy over expressive depth. Emotion recognition and high-level semantic modeling represent another direction. Recent SIGGRAPH work demonstrates that diffusion models can be adapted for interactive texture and brush-pattern generation on 3D surfaces, enabling real-time manipulation of stroke appearance, seamless texture transitions, and locally consistent brush identity, capabilities that point to practical routes for approximating Gongbi’s layered coloration and fine surface detail (Hu et al., 2024). Li et al. (2021) built a multimodal network that classifies emotional content in Chinese figure paintings. While effective, these models rely heavily on dataset availability and may oversimplify culturally embedded emotional vocabularies. In addition to control-based and simulation-based approaches, lightweight text-guided diffusion methods such as FreeStyle demonstrate that stylistic cues can be injected directly through prompts, providing an accessible alternative to full model retraining for aesthetic adaptation (He et al., 2024).
3. Specific genres of painting: Landscape, Bird-and-Flower, and Figure Painting.
The third type of research concerns the selection of research topics:
Landscape and bird-and-flower genres have seen relatively strong results due to their abstract textures and patterned compositions, which align well with current generative models.
Figure painting, however, remains underdeveloped. The scarcity of high-quality datasets, the complexity of facial stylization (e.g., phoenix eyes, cherry lips) and the symbolic meaning of figures varies significantly across different dynasties. Therefore, the generative transformation of figure painting lags behind other subjects, not only due to data scarcity but also because imagery and cultural symbolism are more difficult to encode through statistical modeling.
Recent work on parameter-efficient fine-tuning, particularly LoRA, provides potential solutions by enabling small-scale cultural or stylistic datasets to meaningfully adapt large diffusion models (Zeng et al., 2023). Such as LoRA+, enabled large diffusion models to be effectively tuned on limited-domain artistic corpora without full retraining (Hayou et al., 2024). This offers a feasible technical pathway for adapting pretrained models to the symbolic conventions and stylized morphology characteristic of Chinese figure painting. However, even with LoRA, capturing intentionality, symbolic nuance, and embodied figural expressivity remains a substantial challenge.
4 Recreating the space of in virtual reality
Virtual reality (VR) has evolved from early real-time 3D graphics systems into a mature medium capable of producing immersive, interactive environments (Brooks et al., 1992). As digitization progresses, VR has increasingly been applied to cultural heritage, including. Early efforts focused on database-driven digital preservation (Drew et al., 2017; Sartori, 2016; Stuedahl, 2018), followed by the emergence of virtual museums that employ 3D imaging, interaction techniques, and VR displays to expand accessibility and enhance cultural transmission (Bae et al., 2018; Sirikulpipat and Nadprasert, 2020). For instance, Li and Yu (2020) designed a virtual roaming system for the QAU Museum of Ancient High-Replica Paintings and Calligraphy, demonstrating how VR can transcend spatial constraints, facilitate detailed viewing, and can be preserved for a long time without being easily affected by the environment.
However, research on virtual reality technology and traditional Chinese painting is becoming increasingly diversified, and sometimes even appears fragmented. To clarify the concepts, existing research can be summarized into three major areas: (1) immersive reconstruction of painting space; (2) immersive interaction and experience; and (3) reconstruction of cultural assets related to painting.
4.1 Immersive reconstruction of painting space
A core research direction aims to reconstruct the spatial experience inherent in traditional paintings in a way more acceptable to modern media. Jin et al. (2022) used virtual reality technology to reconstruct this research method, taking the “Spring Dawn in the Han Palace” scroll as an example. Using Unity and Maya software, they transformed the architectural environment and figures’ activities in the scroll into a navigable 3D space. Users can manipulate virtual objects, observe fabric details, and imitate figure movements. Despite these technical achievements, the project also highlights a fundamental aesthetic tension: does transforming the scroll’s “scattered perspective” into the virtual reality’s default focused viewpoint alter the original viewing logic?
Traditional scrolls imbue their space with temporality through the process of being “unfolded” with the act of viewing, emphasizing the viewer’s movement, the intervention of time, and the participation of their inner state. However, virtual reality technology often imposes a unified, immersive perspective. Therefore, virtual reality reconstruction may shift from interpretive to conceptual reconstruction, raising the question: do such works retain the spatial philosophy inherent in traditional Chinese painting, or do they rewrite it?
Landscape reconstruction research further clarifies this contradiction. Zhou et al. (2020) proposed an integrated approach to terrain and water flow modeling to create 3D landscape environments, achieving real-time rendering consistent with traditional painting scenes. These methods expand the experiential dimensions of landscape aesthetics, but their emphasis on visual realism may weaken the essential painterly abstraction in Chinese landscape painting, weaken the painterly abstraction central to traditional landscapes where ink, void, and symbolic brushwork construct an intentionally non-naturalistic space. VR’s emphasis on physical accuracy can inadvertently displace the metaphorical and poetic dimensions that constitute the fundamental mechanisms of yijing construction in traditional Chinese painting.
4.2 Immersive interaction and aesthetic experience
Another research direction explores how virtual reality (VR) reshapes viewers’ cognition, engagement, and aesthetic responses. Yang et al. (2024) combined electroencephalography (EEG) monitoring with VR to study attention during art appreciation. Participants who viewed paintings in VR (and interacted with them using virtual avatars) showed higher aesthetic engagement and post-task attention levels than those using traditional displays. This suggests that VR not only alters perceived immersion but may also reshape people’s cognitive and emotional investment when appreciating traditional art.
Research on memory and perception provides further insights. A study by Krokos et al. (2019) showed that participants using head-mounted virtual reality devices in a “virtual memory palace” had higher memory accuracy than those viewing the same content on a desktop screen. Their findings highlight the advantages of immersive spatialization in memory: perhaps this insight can be applied to designing VR-based scroll or mural exhibitions, as spatial concepts are crucial in these contexts. Similarly, Bamatraf et al. (2016) proposed an EEG-based pattern recognition method to distinguish between real and spurious memories, suggesting that this method could be used to assess cognitive and aesthetic responses in virtual reality museums.
However, VR immersion does not guarantee faithful aesthetic transmission. Recent studies note that VR’s sensory intensity can overstimulate perception and reduce contemplative focus (Mimnaugh et al., 2023), a quality essential to appreciating qiyun and yijing (artistic conception). Further, VR environments often encourage constant movement, interaction, and exploration, which may conflict with the slow, reflective pacing traditionally associated with painting appreciation.
Crucially, VR substitutes the “mental wandering” characteristic of classical Chinese aesthetics with an “embodied roaming” that centers sensorimotor engagement over imaginative participation. This shift raises foundational aesthetic questions: Does VR amplify aesthetic presence or diminish poetic distance? Does immersion enhance understanding or obscure the subtle atmospheric qualities central to traditional Chinese painting?
These studies collectively demonstrate that VR reconstructs aesthetic activity by allowing users to alter viewpoints, engage more actively with cultural symbols, and integrate perception, creativity, and imagination within a digitally mediated environment. This body of work highlights VR’s potential to expand the experiential dimensions of art appreciation.
However, most research foregrounds the advantages of immersion while insufficiently addressing how VR’s sensory intensity may overshadow the atmosphere, yijing and vitality to traditional Chinese painting. This oversight suggests a need for research frameworks that consider not only perceptual stimulation but also the aesthetic and philosophical principles underpinning traditional art.
4.3 Reconstruction of cultural assets related to painting
Several smaller but equally important studies focus on reconstructing cultural elements closely related to the context of painting, particularly the clothing depicted in figure paintings. While these studies do not directly reconstruct the paintings themselves, they help to restore and interpret the socio-cultural environment in which traditional images existed at the time.
For example, Liu et al. (2023) used 3D virtual try-on technology to reconstruct the clothing in “Night Banquet of Xizai of the Han Dynasty”, accurately simulating textile patterns and sewing structures. Zhu et al. (2022) digitally replicated the clothing from “Dou Lian Tu”, dating back over a thousand years, by simulating properties such as the tensile strength and flexural modulus of fabrics. Wang Z. et al. (2024) used Style3D technology to reconstruct the clothing in the “Chao Yuan Tu” mural in Yongle Palace, thus integrating historically accurate clothing into a virtual reality environment.
These reconstructions support research on the aesthetic understanding of figure paintings based on virtual reality (VR) by providing accurate cultural resources, which are crucial for understanding symbolic meaning, social identity, and artistic expression. However, their relevance must be clarified; they should only serve as contextual supplements, not the core of VR painting research. Cultural reconstructions should support, rather than be mistaken for, the core aesthetic logic of traditional Chinese painting. Overemphasis on materials and realism may risk shifting attention from painting’s philosophical and artistic essence.
5 AIGC and VR: a comparative framework for the reconstruction of traditional Chinese painting
Although AIGC and VR are often addressed separately as technological interventions in the digital transformation of Chinese painting, their conceptual orientations, modes of reconstruction, and aesthetic implications reveal complementary yet fundamentally distinct logics. AIGC functions primarily as a productive technology, reconstructing the painterly surface through computational modeling of brushwork, ink diffusion, compositional rhythm, and stylistic attributes. VR, by contrast, operates as an experiential technology, reshaping the spatial, temporal, and perceptual conditions under which painting is viewed. Together, they form two intersecting axes of transformation: one centered on how images are generated, and the other on how images are experienced.
From a technical standpoint, AIGC emphasizes controllability, generativity, and stylistic adaptation. Its central challenges lie in determining whether aesthetic principles such as gong (technical refinement), liubai (productive emptiness), or qiyun can be formalized into computational rules, and whether the implicit intentionality of brush practice can be translated into algorithmic structures. VR, in contrast, foregrounds immersion, spatial reconstruction, and embodied perception. It directly confronts foundational aspects of traditional viewing—such as the sequential unfolding of handscrolls, the contemplative rhythm of wo you (guided wandering), and the temporal-spatial fluidity of shan shui landscapes—by replacing them with a unified, real-time embodied viewpoint.
Aesthetically, the two technologies challenge traditional Chinese painting from opposite directions. AIGC risks reducing painterly meaning to stylistic surface, particularly when diffusion priors or stochastic sampling conflict with the classical order of “idea before brush” (yi xian yu bi). VR, conversely, risks overwhelming subtle atmospheric qualities with sensory intensity. Where AIGC compresses expressive intention into parameterized forms, VR dilates the experiential field and may dissolve the contemplative distance that classical aesthetics intentionally cultivates. Put differently, AIGC struggles to encode intention, while VR struggles to preserve distance.
The cultural implications of these differences are equally significant. AIGC raises questions concerning the fidelity of aesthetic transmission—whether generative models can inherit tacit knowledge embedded in brush technique, poetic imagery, or literati subjectivity. VR, on the other hand, interrogates the phenomenology of viewing: What does it mean to “enter” a painting when entry, in classical aesthetics, is conceived as an intellectual, meditative, and temporally extended process rather than as physical immersion? Can embodied navigation coexist with the reflective detachment central to traditional Chinese painting?
Despite these tensions, AIGC and VR should not be understood as oppositional. Instead, they reveal a potentially productive complementarity. AIGC supplies fine-grained painterly material that VR alone cannot generate, while VR provides spatial and experiential frameworks that contextualize and activate AIGC outputs. Hybrid reconstruction thus becomes possible: algorithmically generated brushwork populating navigable poetic worlds; literati imagery rendered simultaneously computationally expressible and experientially traversable; and traditional aesthetic principles articulated across both surface and space.
Accordingly, a comprehensive digital transformation of Chinese painting requires attending to both aesthetic rules (AIGC) and aesthetic conditions (VR). Only through their integration can digital systems aim not merely to imitate the visual appearance of traditional painting, but to approximate the philosophical, perceptual, and experiential foundations that confer its enduring artistic value.
6 Conclusion
The rapid development of contemporary AIGC and VR technologies has provided a substantial impetus for the digital transformation of. Its impact is not only reflected in changes to creative methods and dissemination paths, but also in a profound restructuring of aesthetic experience, viewing patterns, and image logic. In this sense, “digital transformation” in the context of Chinese painting refers not merely to a shift in medium, but to a deeper reconfiguration of how images are produced, how they are culturally understood, and how viewers experience spatial and temporal relations. Current research shows that digital technology has made significant progress in brushstroke simulation, image style transfer, and virtual scene construction; however, these achievements mainly focus on the reproduction of visual style, while discussions on the core spirit of brush and ink, cultural connotations, and aesthetic mechanisms of traditional Chinese painting remain very limited. This limitation is partly due to the lack of clarification of key aesthetic concepts which are foundational to the evaluative system of traditional Chinese painting and necessary for international readers to understand its visual logic.
This study points out that the unique brushstroke structure, ink tones, and blank space logic of traditional Chinese painting should become important dimensions for evaluating the quality of generated paintings in the future. Such dimensions indicate that future systems must shift from replicating stylistic surfaces to modeling underlying aesthetic mechanisms, a direction increasingly emphasized in recent computational aesthetics research. Although existing AIGC systems mostly use image databases as a basis for style “fitting,” these methods often fail to capture the “flow” and rhythm of brush and ink over time, and also struggle to reflect the aesthetic core of traditional Chinese painting: “vibrant spirit” and “intention preceding the brush.” Addressing this requires computational representations of intentionality, rhythm, and temporal unfolding, rather than static visual features alone. Future technological development needs to move from “style imitation” to “mechanism modeling,” incorporating elements such as brushwork forms, brushstroke patterns, scattered spatial relationships, and the aesthetics of reality and illusion into the generation mechanism itself, rather than treating them as superficial features for later fitting.
Regarding the analogy of technological impact, this study further emphasizes that cameras and AIGC/VR differ fundamentally in their technological essence and the way they impact art. In addition, VR introduces an experiential shift by replacing contemplative, sequential viewing with embodied immersion, raising questions about how digital environments alter the traditional aesthetic distance essential to scroll and landscape painting. Photography, with “recording reality” at its core, primarily influences painting through its substitutability for visual reality and the popularization of image production; while the mechanism of AIGC/VR is to generate new image structures, shape new aesthetic perceptions, and redefine the relationship between creator and work. Thus, while photography challenged representation, AIGC challenges authorship, intentionality, and the epistemic grounds of image-making itself,a distinction that cannot be captured by the analogy that “photography did not replace painting. If photography changes the “way of representing the world,” then AIGC changes the “way of constructing images and meaning.” Therefore, simply comparing “photography has not replaced painting” cannot fully explain the impact of AIGC on traditional painting; a deeper analysis from multiple dimensions, technological characteristics, cultural context, and artistic language,is needed.
Finally, future research directions need to be more specific and actionable. To address the gaps identified above, future work must align technological development with aesthetic, cultural, and phenomenological principles rather than focusing solely on technical performance. This study suggests focusing on the following three directions:
1. Introduce the aesthetic principles and cultural judgment mechanisms of traditional Chinese painting into the AIGC generation logic, enabling the model to move from superficial style imitation to computational expression at the level of “artistic conception-value”. Potential approaches include embedding dynamic stroke-evolution models, qiyun-sensitive evaluators, and generative constraints informed by traditional painting theory;
2. Systematically explore the structural differences between VR and traditional scattered perspective and travel-style viewing, develop virtual presentation methods adapted to the spatial logic of traditional Chinese painting, and avoid weakening traditional aesthetics by a single perspective system. This includes VR modes that simulate “scroll-like” temporal unfolding, multi-perspective roaming consistent with scattered perspective, and experience designs that support contemplative rather than overstimulated viewing;
3. Deepen the participation mechanism of artists in digital creation, construct a “human-machine co-creation” process with cultural subjectivity, and ensure that digital transformation does not damage the spiritual core of traditional Chinese painting while innovating. Such forms of collaboration ensure that algorithmic systems remain guided by human intentionality and culturally grounded judgment, rather than replacing them.
Author contributions
HC: Writing – original draft, Writing – review and editing.
Funding
The author(s) declared that financial support was not received for this work and/or its publication.
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher’s note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Bae, J. A., Mu, S., Kim, J. S., Turner, N. L., Tartavull, I., Kemnitz, N., et al. (2018). Digital museum of retinal ganglion cells with dense anatomy and physiology. Cell 173 (5), 1293–1306. doi:10.1016/j.cell.2018.04.040
Bamatraf, S., Hussain, M., Aboalsamh, H., Qazi, E. U. H., Malik, A. S., Amin, H. U., et al. (2016). A system for true and false memory prediction based on 2d and 3d educational contents and EEG brain signals. Comput. Intell. Neurosci. 2016 (1), 8491046. doi:10.1155/2016/8491046
Brooks, J. , F., Burbeck, C., Durlach, N., Ellis, M. S., Lackner, J., Robinett, W., et al. (1992). Research directions in virtual environments. Comput. Graph. 26 (3), 153–177. doi:10.1145/142413.142416
Chu, N. S. H., and Tai, C.-L. (2004). Real-Time painting with an expressive virtual Chinese brush. IEEE Comput. Graph. Appl. 24 (5), 76–85. doi:10.1109/mcg.2004.37
Chung, J., Kim, S., Lee, H., and Park, J. (2023). Training-free artistic style injection for diffusion models. arXiv:2310.06709.
Drew, J. A., Moreau, C. S., and Stiassny, M. L. (2017). Digitization of museum collections holds the potential to enhance researcher diversity. Nat. Ecology & Evolution 1 (12), 1789–1790. doi:10.1038/s41559-017-0401-6
Grant, N., and Metz, C. (2022). A new chat bot is a'Code red'for google's search business. International New York Times.
Hayou, S., Ghosh, N., and Yu, B. (2024). Lora+: efficient low rank adaptation of large models. arXiv preprint arXiv:2402.12354.
He, B., Gao, F., Ma, D., Shi, B., and Duan, L.-Y. (2018). “ChipGAN,” in Proceedings of the 26th ACM international conference on multimedia.
He, F., Li, G., Zhang, M., Yan, L., Si, L., Li, F., et al. (2024). Freestyle: free lunch for text-guided style transfer using diffusion models. arXiv preprint arXiv:2401.15636.
Hu, A., Desai, N., Abu Alhaija, H., Kim, S. W., and Shugrina, M. (2024). “Diffusion texture painting,” in ACM SIGGRAPH 2024 conference papers, 1–12.
Jin, S., Fan, M., and Kadir, A. (2022). Immersive spring morning in the Han Palac e: learning traditional Chinese art via virtual reality and multi-touch tabletop. Int. J. Human-Computer Interact. 38 (3), 213–226. doi:10.1080/10447318.2021.1930389
Krokos, E., Plaisant, C., and Varshney, A. (2019). Virtual memory palaces: immersion aids recall. Virtual Reality 23 (1), 1–15. doi:10.1007/s10055-018-0346-3
Lee, L. W., Dabirian, A., McCarthy, I. P., and Kietzmann, J. (2020). Making sense of text: artificial intelligence-enabled content analysis. Eur. J. Mark. 54 (3), 615–644. doi:10.1108/ejm-02-2019-0219
Li, J., and Yu, N. (2020). Key technology of virtual roaming system in the museum of ancient high-imitative calligraphy and paintings. IEEE Access 8, 151072–151086. doi:10.1109/access.2020.3015318
Li, J., Chen, D., Yu, N., and Zhao, Z. (2021). Emotion recognition oftraditional Chinese paintings at the thirteenth national exhibition of fines arts in China based on advanced affective computing. Front. Psychol. 12 (15).doi:10.3389/fpsyg.2021.741665
Liu, K., Wu, H., Gao, Y., Zhu, C., Ji, Y., and Lü, Z. (2023). Archaeology and virtual simulation restoration of costumes in the Han Xizai banquet painting. Autex Res. J. 23 (2), 238–252. doi:10.2478/aut-2022-0001
Mimnaugh, K. J., Center, E. G., Suomalainen, M., Becerra, I., Lozano, E., Murrieta-Cid, R., et al. (2023). Virtual reality sickness reduces attention during immersive experiences. IEEE Trans. Vis. Comput. Graph. 29 (11), 4394–4404. doi:10.1109/TVCG.2023.3320222
Sartori, A. (2016). Towards an intellectual history of digitization: myths, dystopias, and discursive shifts in museum computing. Digital Scholarsh. Humanit. 31 (2), 428–440. doi:10.1093/llc/fqv006
Sheng, J., Song, C., Wang, J., and Han, Y. (2019). Convolutional neural network style transfer towardstraditional Chinese paintings. IEEE Access 7 (7), 163719–163728. doi:10.1109/access.2019.2952616
Sirikulpipat, P., and Nadprasert, P. (2020). Development virtual museum interactive application with 360-degree and 3D scans virtual reality technology: case study 10 Museums of Thailand. J. Inf. Sci. Res. Pract. 38 (1), 42–58. doi:10.14456/jiskku.2020.3
Stuedahl, D. (2018). “Participation in design and changing practices of museum development,” in The Routledge handbook of museums, media and communication (London, UK: Routledge), 219–231.
Wang, B., Chen, Q., and Wang, Z. (2025). Diffusion-based visual art creation: a survey and new perspectives. ACM Comput. Surv. 57 (10), 1–37. doi:10.1145/3728459
Wang, X., Huo, B., Yuan, X., Lu, Z., Liu, F., Zhou, S., et al. (2024). Bringing the queen mother of the west to life: digital reconstruction and analysis of Taoist Celestial Beings Worshiping mural’s apparel. AUTEX Res. J. 24 (1), 20230047. doi:10.1515/aut-2023-0047
Wang, X., Du, Y., and Pang, Y. (2025). Inkartgan: a deep learning-based approach for digital generation of ink painting. J. Intelligent Manuf. doi:10.1007/s10845-025-02674-6
Wang, Z., Liu, F., and Ran, C. (2024). Cvstgan: a controllable generative adversarial network for video style transfer oftraditional Chinese painting. Multimed. Syst. 30 (5). doi:10.1007/s00530-024-01457-y
Wong, H. T. F., and Ip, H. H. S. (2000). Virtual brush: a model-based synthesis of Chinese calligraphy. Comput. & Graph. 24 (1), 99–113. doi:10.1016/s0097-8493(99)00141-7
Yang, X., Cheng, P. Y., Liu, X., and Shih, S. P. (2024). The impact of immersive virtual reality on art education: a study of flow state, cognitive load, brain state, and motivation. Educ. Inf. Technol. 29 (5), 6087–6106. doi:10.1007/s10639-023-12041-8
Zhang, L., Rao, A., and Agrawala, M. (2023). “Adding conditional control to text-to-image diffusion models,” in Proceedings of the IEEE/CVF international conference on computer vision, 3836–3847.
Zhou, P., Li, K., Wei, W., Wang, Z., and Zhou, M. (2020). Fast generation method of 3D scene in Chinese landscape painting. Multimedia Tools Appl. 79 (23), 16441–16457. doi:10.1007/s11042-019-7476-9
Zhu, C., Liu, K., Li, X., Zeng, Q., Wang, R., Zhang, B., et al. (2022). Research on archaeology and digital restoration of costumes in DaoLian painting. Sustainability 14 (21), 14054. doi:10.3390/su142114054
Keywords: AIGC, artificial intelligence, brush-and-ink aesthetics, computational aesthetics, cultural computing, digital art wave, digital heritage, scattered perspective
Citation: Cao H (2026) The transformation of traditional Chinese painting in the digital art wave: the impact of AIGC and virtual reality. Front. Virtual Real. 6:1716174. doi: 10.3389/frvir.2025.1716174
Received: 08 October 2025; Accepted: 19 December 2025;
Published: 21 January 2026.
Edited by:
Eunju Hwang, University College London, United KingdomReviewed by:
Vincenzo De Masi, Guangdong University of Foreign Studies, ChinaJiufang Lv, Nanjing Forestry University, China
Copyright © 2026 Cao. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Hui Cao, Y2FvaHVpNzAzQG91dGxvb2suY29t