Adobe Certified Professional - Adobe Firefly flashcards
Adobe Certified Professional - Adobe Firefly · Build recall with a focused review session.
32 cards. Read the term and the definition here. The deck at the top of this page is optional.
- Generative AI
- A class of models that learn patterns from training data and generate new content such as images, text, audio, or video in response to prompts.
- Adobe Firefly
- Adobe's family of generative AI models and features integrated across Creative Cloud, designed for commercial use and trained on licensed and public-domain content.
- Prompt
- The text instruction given to a generative model describing the desired output; its specificity strongly influences the generated result.
- Seed
- A numeric value that initializes the model's random process; keeping the same seed and parameters helps reproduce a generation.
- Generative credits
- Units consumed when using Firefly features; different models and operations cost varying amounts, and plans include monthly allocations.
- Inpainting
- A generative editing technique that fills a selected region of an image with synthesized content based on a prompt and surrounding context.
- Outpainting
- A generative technique that extends an image beyond its original boundaries, synthesizing new pixels that continue the existing scene.
- Generative Fill
- A Photoshop and Firefly feature that synthesizes content inside a user-defined selection using a text prompt and surrounding image context.
- Generative Expand
- A Firefly and Photoshop feature that enlarges the canvas and generates new content to fill the extended area, continuing the original image.
- Generative layer
- A non-destructive Photoshop layer holding a Firefly generation; its prompt and settings can be revisited and regenerated without altering base pixels.
- Aspect ratio
- The proportional relationship between an image's width and height, set before generation; it shapes composition and framing of output.
- Negative prompt
- Text describing elements to exclude or suppress in a generation; effectiveness depends on the model version and phrasing.
- Reference image
- An input image used to guide style, composition, or subject of a generation; strength controls how much it influences output.
- Style preset
- A predefined visual treatment (e.g., photographic, illustration, cinematic) that biases a generation toward a particular look.
- Iterative prompting
- The practice of generating, reviewing, and revising prompts or parameters repeatedly to converge on a desired output.
- Text-to-video
- A generative capability that produces a short video clip from a text prompt, with parameters for duration, motion, and aspect ratio.
- Image-to-video
- A generative technique that animates a still image into a short video clip, using the image as the starting frame and prompt for motion.
- Text-to-vector
- A Firefly capability that generates editable vector artwork from a prompt, usable in Illustrator and design workflows.
- Generative recolor
- A feature that applies new color palettes to vector artwork while preserving its structure, shapes, and grouping.
- Generative audio
- AI-generated sound, ambience, music, or speech produced from prompts or text, usable in video and design workflows.
- Content Credentials
- Tamper-evident provenance metadata based on the C2PA standard that records an asset's origin, AI generation, and editing history.
- C2PA
- Coalition for Content Provenance and Authenticity; the open standard underlying Content Credentials for signed provenance metadata.
- Commercial safety
- Adobe's design goal that Firefly outputs are intended for commercial use, supported by training on licensed and public-domain content.
- AI disclosure
- Informing audiences that content was AI-generated or edited; may be required by law, platform policy, or contract.
- Deepfake
- Synthetic media depicting real people or events deceptively; Adobe guidelines prohibit using Firefly to create deceptive deepfakes.
- Human authorship
- The creative human contribution required for copyright protection in many jurisdictions; purely AI-generated output may not be protectable.
- Model version
- A specific release of a generative model; different versions can produce different styles, quality, and prompt adherence.
- Generative upscaling
- Using AI to increase image resolution and synthesize plausible detail; it cannot recover information absent from the source.
- Selection feathering
- Softening the edge of a selection so generated content blends smoothly with surrounding pixels during inpainting.
- Camera movement prompt
- Text describing motion such as pan, zoom, or orbit that guides the animation of a generated video clip.
- Motion strength
- A video generation parameter controlling how much movement occurs in the clip; higher values produce more dynamic motion.
- Firefly web app
- A browser-based interface for Firefly generation and editing, providing access to text-to-image, video, and design features.