17 Types of Digital Art – Examples, Styles & Tools

Types of digital art are ways of making or presenting art with digital technology. They include two- and three-dimensional digital painting, vector and raster illustration, pixel art, CGI and fractal art, alongside VR painting and mixed media, photobashing and algorithmic art.

What Is Digital Art?

Digital art is a term used to describe art that is made or presented using digital technology. It can be computer generated, scanned or drawn using a tablet and a mouse. A related name is multi-media, an artwork that combines electronic media such as video, film, audio and computers, as distinct from mixed media, a combination of different media or materials.

Tate dates the first use of the term digital art to the early 1980s and links it to the paint program used by Harold Cohen that became known as AARON. The Whitney Museum of American Art dates AARON itself to the late 1960s, when Cohen conceived it at the University of California, San Diego, naming it in the early 1970s. Frieder Nake’s university-hosted account documents computer-art exhibitions in 1965 by A. Michael Noll in New York and by Nake and Georg Nees in Germany.

A. Different Types of Digital Art

1. 2D Digital Painting

Two-dimensional digital painting is painting on a flat digital canvas, using tools such as a tablet or a mouse. The Philippine Department of Education’s May 2016 Arts curriculum lists it for Grade 6 with GIMP, an open source program, and Paint on Windows, and the first of its learning competencies for that quarter is that art processes, elements and principles still apply even with the use of technologies.

ArtRage Vitae, for Windows and macOS, offers oils, watercolors and pencils, with layers, blend modes, references, guides, grids and perspective tools. Corel describes Painter as professional digital art software for Windows and Mac with Natural-Media brushes that emulate traditional media.

Procreate provides drawing tools on the iPad, alongside its maker’s Procreate Pocket for the iPhone and Procreate Dreams for animation, and Krita is a professional free and open source painting program made by artists who want affordable art tools for everyone.

2. 3D Digital Painting and Sculpting

3D digital painting and sculpting combines mesh shaping with surface painting. Blender’s sculpting tools include multires sculpting, mirrored sculpting and dynamic topology, or dyntopo, a sculpting method that adds and removes details on the fly, whereas regular sculpting only affects the shape of a mesh, with built-in brushes such as Crease, Clay Strips, Pinch, Grab, Smooth and Mask. Blender is free and open source software, a community project coordinated by the Blender Foundation under the GNU General Public License.

3. Data-moshing

Data-moshing, or datamoshing, is an abstract form of digital art made by breaking a compressed video file on purpose. One method is to open the video file in a binary editor and remove the code that defines an I-frame as the reference for B-frames and P-frames, so the playback software, unable to predict correct images, plays a video substantially different from the original. Datamoshing belongs to glitch art, a trend in media art that uses noise and error for artistic effect.

The three letters name possible frame types used by a video encoder in MPEG-2 and H.264 groups of pictures, I-frames or intra-coded frames, P-frames or predictive frames and B-frames or bidirectional predictive frames. The effect can smear images across frames, producing a painterly quality.

Takeshi Murata’s Monster Movie is a single-channel video of 4 minutes 19 seconds held by the Smithsonian American Art Museum, for which Murata sourced video from the 1981 B movie Caveman and, beginning with a process called datamoshing, mixed the video’s compression into a kind of digital liquid.

Google’s DeepDream, released by Google Research in July 2015, is a different technique. It feeds an image to a trained neural network and asks it to enhance whatever a chosen layer detected, so it belongs to neural-network imaging, not video compression.

4. Pixel Art

Pixel art builds an image pixel by pixel on a visible grid, and the name was in print by December 1982, when Goldberg and Flegal published an ACM president’s letter titled Pixel Art in Communications of the ACM. Video games carry much of its history, and the 14 video games the Museum of Modern Art, MoMA, acquired in 2012 for its Architecture and Design collection included Pac-Man of 1980 and Tetris of 1984.

Aseprite is an animated sprite editor and pixel art tool. Its tools include Pixel Perfect Stroke and RotSprite rotation, designed to reduce pixel distortion when rotating tiny sprites. Finished work exports as animated GIFs or as sprite sheets in PNG and JSON files.

5. Algorithmic Art

Algorithmic art can be made by writing a program that directs a computer’s drawing, as Frieder Nake did. Frieder Nake used the ZUSE Z64 Graphomat at the Technical University of Stuttgart, an early computer-driven drawing machine, and his Hommage à Paul Klee 13/9/65 Nr.2 of 1965 turns a literal analysis of Klee’s 1929 oil painting Highroads and Byroads into an algorithm. Nake deliberately wrote random variables into the program, so chance and rule shape the drawing together.

Vera Molnár began working with the computer in Paris in 1968, and the V&A holds her Letters from my Mother of 1988, a screenprint after plotter drawings that simulate her mother’s handwriting. For writing such programs today, Processing, free and open source since its first release on August 9, 2001, was created as a software sketchbook to teach programming within a visual context and is used in art schools and visual arts programs.

6. Fractal Art

Fractal art is algorithmic art built on fractal geometry, the field of the mathematician Benoit Mandelbrot, who spent 35 years at IBM and was among the first to use computer graphics to create and display complex fractal geometric images. Worcester Polytechnic Institute dates Mandelbrot’s coinage of fractal to 1975; the University of St Andrews records its appearance in his 1975 book but also cites Johnson’s attribution to a 1976 article (both sources read October 22, 2026). The term describes irregular shapes in nature and mathematics that exhibit self-similarity, so a pattern repeats as the view zooms in, and comes from the Latin fractus, or broken.

Mandelbrot set out his fractal geometry in Les objets fractals of 1975 and more fully in The Fractal Geometry of Nature of 1982, which brought his ideas widespread attention. Ultra Fractal for Windows and macOS offers thousands of fractal types and coloring algorithms, zoom controls, gradients to add color and multiple layers to combine different fractals in one image.

7. Still Imagery

Still imagery is a static visual representation, such as a painting, drawing, graphic design, plan or map. In the Type Vocabulary of the Dublin Core Metadata Initiative, an instance of Still Image also belongs to the broader resource type Image.

Digital stills can use raster images built from pixels or vector graphics built from lines and curves. Adobe announced in July 2017 that updates and distribution of Flash Player would end in 2020. For vector stills, Adobe Illustrator is a vector-based program that specializes in creating scalable vector graphics, logos and design elements.

8. CGI Art

CGI art, or computer-generated imagery, can be two- or three-dimensional, static or animated. Its landmark is Toy Story, released in theaters on November 22, 1995, the world’s first computer animated feature film. Pixar’s history credits the Lucasfilm Computer Division’s Genesis Effect for Star Trek II The Wrath of Khan as the first completely computer animated sequence in a feature film. The same modeling, sculpting and rendering work can be done in Blender, where a character can be rigged with an armature, a type of object that works like a real skeleton, whose bones move and deform whatever is attached to them. Rendering of this kind also serves architectural visualization, where Chaos offers V-Ray and Corona, built on physically based rendering with accurate lighting, materials and cameras.

9. Vector Painting

Vector painting builds shapes from mathematical equations, lines and curves that remain sharp when scaled. Adobe’s general raster-versus-vector comparison describes vector files as containing no pixels; its Illustrator SVG export documentation also allows raster images to be embedded in the file (both read October 22, 2026). A brochure can combine vector graphics for its company logo with raster photographs. Inkscape is a free and open source vector graphics editor that uses SVG as its main file format.

10. Raster Painting

Raster painting stores every stroke, blend and texture as pixels, and enlargement can make a raster image look blurred or pixelated. A bitmap is a computer image formed from pixels that are each stored as a value of one or more bits, and the Bitmap Image File, BMP, is one of the raster file types. Painting programs for this work include Krita and ArtRage Vitae. Adobe Photoshop is a raster-based program, meaning it uses pixels to create detailed images.

11. VR Painting

VR painting puts the brush strokes in three-dimensional space around the artist, and one tool for it, Tilt Brush, a painting app, launched on SteamVR for the HTC Vive in April 2016 and ran on the Oculus Rift, Windows Mixed Reality, Valve Index, PlayStation VR and Oculus Quest headsets.

Google ended active development and released the source code on January 26, 2021, following version 23 in March 2020, and the work continues as Open Brush, a community-led evolution of Tilt Brush, a room-scale 3D painting XR application, now available for free.

12. Photobashing

Photobashing combines found photographs with digital painting to create a new image. The Gnomon Workshop’s cinematic concept-art course for films and games moves from sketching and photobashing in Photoshop to sculpting, modeling, materials and lighting in Blender for the entertainment industry.

In our view, Surrealism can be a reference point for dreamlike photobashed scenes. Guillaume Apollinaire coined the word surrealist in 1917 before André Breton defined the movement in his Surrealist Manifesto of 1924, and Salvador Dalí, 1904 to 1989, is among the movement’s artists.

13. Photo Painting

Photo painting uses a photograph as the base for a digitally painted image, with digital brushwork applied over the photographic forms. Corel Painter provides photo-art tools alongside its blank-canvas painting tools and its Natural-Media brushes for emulating traditional painting media.

14. Digital Collage

Digital collage is the screen version of collage, both the technique and the resulting work in which pieces of paper, photographs, fabric and other ephemera are arranged and stuck down onto a supporting surface. Tate derives the term from the French papiers collés and places its first use as an artists’ technique in the early twentieth century. National Galleries of Scotland gives a longer history: collage techniques existed for centuries before Picasso and Braque adopted the term (both sources read October 22, 2026). Separate image layers let a digital collage’s elements be repositioned and edited individually.

Tate describes artists downloading video onto computers in the 1990s and cutting and pasting within moving images to create visual collages. On the consumer side, Google retired its Picasa photo program to focus on Google Photos and stopped supporting the Picasa desktop application on March 15, 2016.

15. Integrated Art

We use integrated art as a working label for pieces that combine different art forms or media, such as painting with photography, flat with three-dimensional graphics or vector with raster work, a practical grouping for discussing these combinations. Multi-media combines electronic media.

16. Mixed Media

Mixed media describes artworks composed from a combination of different media or materials, and in digital work each medium, from paint to photographs and three-dimensional models, is brought into one piece. Tate traces mixed media in modern art to around 1912 and the cubist collages and constructions of Pablo Picasso and Georges Braque.

17. Digital Photography

Digital photography records the image on an electronic sensor and stores it as data. In 1975 Steven Sasson, an engineer at Eastman Kodak, built the first hand-held solid-state still camera, now an IEEE Milestone, a self-contained electronic camera using a Fairchild 100 x 100 pixel charge-coupled device image sensor, movie camera optics, a temporary RAM buffer and storage on a digital cassette. IEEE’s milestone proposal identifies the sensor as a Fairchild MV-101, whereas IEEE Spectrum’s July 2020 account identifies it as type 201; the Digital Camera Museum identifies MV-101 as a camera model (all read October 22, 2026).

Famous Examples of Digital Art

Famous examples of digital art include Vera Molnár’s Letters from my Mother, Takeshi Murata’s Monster Movie and Pac-Man in MoMA’s collection. Museums and other institutions hold, sell or document these works, preserving their dates alongside the names of their artists and makers.

What Are the Main Digital Art Styles?

Our working distinction is that a type names how a piece is made, from raster painting to datamoshing, while a style names how it looks. We find this useful for comparing works that share a technique but pursue different visual effects.

Blender’s Freestyle is an edge and line-based non-photorealistic rendering engine whose line styles produce artistic looks, such as hand drawn and painted, or technical hard-line looks, with pre-scripted styles such as Japanese big brush, cartoon and blueprint. Its Cycles-only Toon BSDF shader creates diffuse and glossy materials with cartoon light effects, so a cartoon style is a rendering choice inside the same 3D and CGI type.

Technology-Based Art in the DepEd Grade 10 Curriculum

In the Philippines, the Department of Education’s K to 12 Arts Curriculum Guide of May 2016 sets Technology-Based Art as the Grade 10 second-quarter content. Under Computer and Digital Arts it lists cellular phones for photos and videos, computer-generated images, digital photography with DSLR and point-and-shoot cameras, video games, digital painting and imaging, and videos for TV and film. One of its learning competencies for the quarter is to apply different media techniques and processes showing the characteristics of 21st-century art, for example with graphic software like Photoshop and InDesign.

The matching Grade 10 Quarter 2 self-learning modules are Elements and Distinct Characteristics of Technology-Based Production Arts, Technology-Based Artworks with Their Role, Function and Characteristics, and Digital or Computer Arts Application, released on DepEd’s learning resource portal on August 3, 2022.

B. AI Art

AI art has a history that runs from Harold Cohen’s AARON to the text-to-image systems and the US Copyright Office’s 2025 report on their output.

• In the late 1960s Harold Cohen conceived AARON, the earliest artificial intelligence program for artmaking, at the University of California, San Diego, and he named it in the early 1970s.
• On June 18, 2015 Google Research described its Inceptionism technique, and on July 1, 2015 it released the code as DeepDream after finding that the process also generated art.
• On 25 October 2018 Portrait of Edmond de Belamy, created by Obvious with a generative adversarial network, sold in Christie’s Prints & Multiples sale in New York. Christie’s announcement of August 20, 2018 called it the first major auction house to offer algorithm-created art; its feature dated December 12, 2018 used the broader first auction house claim.
• In January 2021 OpenAI introduced DALL-E, and in 2022, DALL-E 2.
• In 2022 Jason M. Allen’s Théâtre D’opéra Spatial, made with Midjourney, won first place in Digital Arts / Digitally-Manipulated Photography in the Colorado State Fair’s Emerging Artist division.
• On September 5, 2023 the US Copyright Office Review Board affirmed the refusal to register Théâtre D’opéra Spatial as submitted.
• In 2023 OpenAI followed with DALL-E 3, an image generation system built natively on ChatGPT.
• On January 29, 2025 the US Copyright Office published Part 2 of its Copyright and Artificial Intelligence report.

1. DALL-E Artificial Intelligence Digital Art System

Overview

DALL-E was the text-to-image system OpenAI introduced in January 2021, followed in 2022 by DALL-E 2, a system for generating images from natural-language descriptions. OpenAI described DALL-E 2 as producing more realistic and accurate images with four times the resolution of the first system. DALL-E 3 followed in 2023, built natively on ChatGPT. OpenAI removed the DALL-E 2 and DALL-E 3 model snapshots from its API on May 12, 2026.

How Does It Work?

DALL-E generated images from natural-language descriptions, the task OpenAI introduced on January 5, 2021. DALL-E 2 also offered editing within generated or uploaded images and outpainting beyond their borders, which OpenAI described as taking existing shadows, reflections and textures into account.

DALL-E 2 was built on CLIP, the model that links text with images, and the research describing its image-generation method was published on April 13, 2022 under the title Hierarchical text-conditional image generation with CLIP latents, identifying the method’s connection to that model.

Drawback

AI-generated output raises questions about copyright protection. The US Copyright Office concluded in January 2025 that the outputs of generative AI can be protected by copyright only where a human author has determined sufficient expressive elements, such as a perceptible human-authored work in the output or creative arrangement or modification of it, and the mere provision of prompts does not qualify.

AI assistance does not by itself bar copyrightability. A human can use AI during the process of creation or include AI-generated material within a larger human-generated work, and the position described by the Copyright Office applies to US law only.

2. Midjourney AI Digital Art System

Overview

Midjourney is a text-to-picture artificial intelligence service that turns a prompt into an image. Like DALL-E, it takes a short description written in words by the user. Midjourney describes its /imagine command as generating a unique image from a short text prompt.

How Does It Work?

On Discord, a subscription is needed before creating images with Midjourney. The /imagine command takes the prompt and returns an initial image grid, and two rows of buttons then appear underneath the grid to pick or vary an image. In earlier versions of Midjourney the U buttons were used for upscaling images. Parameters added to the prompt control aspects of the output, such as the image’s aspect ratio or how much computing time is spent on it. Getting a finished image can take many rounds, and Jason M. Allen, by his own count, input numerous revisions and text prompts at least 624 times to arrive at the initial version of Théâtre D’opéra Spatial.

Drawback

Midjourney output faces the same US copyright requirement for human authorship. In its decision of September 5, 2023 the US Copyright Office Review Board found that the Midjourney image, which remains in substantial form in Allen’s final work, is not the product of human authorship, and that the AI-generated content must be disclaimed. Part of the art community also rejects the tools, Procreate under the banner No generative AI and the line Creativity is made, not generated.

Digital Art Final Thoughts

Digital art runs from the computer art that A. Michael Noll, Frieder Nake and Georg Nees exhibited in 1965 to today’s text-prompted image systems, and includes computer-generated, scanned and digitally drawn work, as Tate describes. On US copyright protection, the Copyright Office concluded in January 2025 that prompts alone do not establish authorship of generated output, and painting apps such as Procreate state No generative AI for their own tools.