DeepDream

ディープドリーム / 2015– / Technique / Digital Art

Google's visualization technique repeatedly amplifies features learned by an image-recognition network, making eyes, animals and towers emerge excessively from clouds, rocks and leaves until misrecognition becomes a recursive hallucinatory surface.

DeepDream — DeepDream applied to white noise, published algorithm-generated example — Martin Thoma, 2016 — MartinThoma — Wikimedia Commons / CC0 — DeepDream applied to white noise, published algorithm-generated example — Martin Thoma, 2016 — MartinThoma — Wikimedia Commons / CC0 · CC0

DeepDream — AI-generated interpretation 1 — AI-generated interpretation, not a reference work

DeepDream — AI-generated interpretation 2 — AI-generated interpretation, not a reference work

At a glance

Visual cues

Eyes and animal faces repeated across the frame
Swirling texture multiplying along source contours
Multi-scale repetition that reveals new figures when viewed closer
Saturated blue, yellow and violet rolling continuously

Best used for

Exhibits and critical images that make machine misrecognition immediately visible
Music video and cover imagery that keeps a source landscape beneath a multi-scale hallucination

Use this style

Take the compact brief into a project or hand the structured version to an AI tool.

design.md image prompt JSON API MCP & machine access Compose

open in ChatGPT open in Claude open in Google AI

Detailed guidance and sourcesProduction rules, reference works, comparisons, research and citation formats

Dictionary entry

Type: Add type after processing, keeping it short, bold and simple enough not to fight the detail.
Composition: Preserve the source's large composition and amplify related features at several scales to create recursion into depth.
Material: Choose layers in a trained vision model and iteratively increase their activation through gradient ascent.
Caution: It is not a swirl filter or pasted dog faces. Preserve which layer and feature were amplified and how they bind to the source.
Further study: Feature visualization and layer selection / Iterative zoom and octave scales / GoogLeNet ImageNet feature bias

Related entries

Generative Art (ジェネラティブアート)Same Digital Art lineage.Compare DeepDream and Generative Art
Psychedelic (サイケデリック)Shared effects: Exhilaration.Compare DeepDream and Psychedelic

Cite this entry

The dictionary is meant to be linked to. Take whichever form fits where you are writing.

Link: https://indexstyle.org/styles/deepdream
Markdown: [DeepDream — IndexStyle](https://indexstyle.org/styles/deepdream)
HTML: <a href="https://indexstyle.org/styles/deepdream">DeepDream — IndexStyle</a>

Back to index position