In the evolution of computer-generated imagery (CGI) and digital art, the notion of texture has expanded beyond mere surface realism to encompass questions of materiality, tactility, and aesthetic expression. While early digital imaging prioritized photorealistic accuracy, recent developments in both digital painting and 3D animation increasingly foreground the handmade look, an aesthetic that emulates the imperfections and material depth of traditional media. Software such as Procreate, Blender, and Substance Painter employ textured brushes and overlay systems that simulate pigment layering, paper grain, and brush pressure, bridging the sensory qualities of painting with the procedural logic of computation. In the realm of animation, stylized works such as Puss in Boots: The Last Wish (2022), Arcane (2021), and Klaus (2019) exemplify a broader artistic shift: the pursuit of painterly authenticity within the digital domain.
Carinna Parraman’s study The Material Image: Artists’ Approaches to Reproducing Texture in Art offers a foundational perspective on this topic. She investigates how artists replicate tactile qualities like grain, gloss, translucency, and relief across analogue and digital platforms. According to Parraman, the key distinction between digital and material art lies in the absence of sensory feedback: “computers have no capability to compare whether a textural rendering looks right or wrong. Only humans can make the final subjective decision.” Texture, in this sense, becomes an epistemological bridge between perception and representation. Digital systems can reproduce surface complexity through code, but their understanding of “material rightness” depends entirely on human judgment. Parraman identifies three perceptual parameters essential to the illusion of texture: value, repetition, and edge. These determine how contrast, pattern, and boundary cues simulate the experience of tactility. In digital tools, these attributes are algorithmically mapped onto layers of colour data and procedural noise. The texture brushes in Procreate, for example, operate by randomizing micro-patterns within controlled statistical limits, producing the illusion of roughness or fibrous density. Similarly, in 3D rendering, shader nodes mimic micro-surface irregularities through bump, normal, and displacement maps and transform optical data into perceived material depth. (Parraman, 2013)
In the context of moving images, texture must also maintain coherence across time. Bénard et al.’s Stylizing Animation by Example is a study that extends Aaron Hertzmann’s “Image Analogies” framework into the temporal domain. Bénard and colleagues address a core difficulty in achieving painterly CGI: maintaining temporal coherence. The consistency of texture patterns, brush marks, and tonal variation from frame to frame. Without such coherence, painterly or textured animations tend to “flicker” or “pop,” breaking the illusion of continuous handcrafted motion. The researchers propose a Temporally Coherent Image Analogies (TCIA) algorithm that applies example-based stylization to animation sequences. In this system, an artist first paints a set of keyframes using digital or traditional media. The algorithm then analyses the relationship of its texture statistics, colour distribution, and spatial correspondences between the base render and the painted keyframe and synthesizes the in-between frames automatically. As Bénard et al. explain, “First, we extend image analogies to animation, achieving temporal continuity while taking account of occlusion and disocclusion. Second, we allow art direction by modifying the approach to interpolate hand-painted elements at keyframes.” By blending procedural texture synthesis with direct artistic input, the method balances computational precision with expressive variability. This approach situates painterly 3D animation within the broader field of non-photorealistic rendering (NPR), where visual style becomes a primary narrative device rather than a byproduct of realism. (Bénard et al., 2013)
The TCIA algorithm’s treatment of texture mirrors Parraman’s notion of perceptual negotiation: while machines handle pattern continuity, the human artist supplies the qualitative cues – brush rhythm, tonal hierarchy, and stroke irregularity. In practice, this combination allows contemporary studios to create animated films that feel “painted,” without the prohibitive labour of frame-by-frame artistry.
The aesthetic implications of Bénard et al.’s system resonate strongly in modern animation pipelines. For instance, Puss in Boots: The Last Wish employs similar non-photorealistic rendering techniques to emulate painterly brushwork. The textures do not strive for photographic realism but instead highlight visible strokes, colour layering, and diffuse light falloff, recalling gouache and pastel illustrations. Each frame becomes a dynamic painting.
This dialogue between the human and the algorithmic highlights a cultural desire to reintroduce the human trace within technologically mediated imagery. Digital texture, whether created through a stylized shader or an example-based synthesis algorithm, becomes a signifier of authorship. It gestures toward the handmade, even when it is entirely code-generated. In films like Arcane and Klaus, textured brushstrokes and light diffusion serve not only as stylistic embellishments but as affective markers that connect viewers to material traditions of painting and illustration.
References
Adelson, E. H. (2001). On seeing stuff: the perception of materials by humans and machines. Proceedings Of SPIE, The International Society For Optical Engineering. https://doi.org/10.1117/12.429489
Bénard, P., Cole, F., Kass, M., Mordatch, I., Hegarty, J., Senn, M. S., Fleischer, K., Pesare, D. & Breeden, K. (2013). Stylizing animation by example. ACM Transactions On Graphics, 32(4), 1–12. https://doi.org/10.1145/2461912.2461929
Parraman, C. (2013). Reproduction of Texture in Digitally Printed Artworks. International Colour Society (AIC) Congress at The Sage Gateshead, (July 2013).