AI video and image glossary
Plain definitions of the words behind AI video and image tools. Each entry says what the term means, how it works, where it breaks and what DreamBench does with it, with diagrams of its own.
AI video generation
Character consistency
Keeping a character recognizably the same person in every shot of an AI video, not just within one clip.
Read the entryFirst and last frame
AI video that opens on one picture and ends on another, with the motion between them generated.
Read the entryImage to video
AI video that starts from a still picture, the motion generated from a prompt.
Read the entryLip sync
A mouth moving in time with speech: made with the voice from a prompt, or redrawn to fit an audio track.
Read the entryMulti-shot video
An AI video of several shots joined by cuts: planned as a shot list, generated shot by shot, cut into one film.
Read the entryNative audio
Sound a video model makes with the picture: dialogue, effects and ambience on one timeline.
Read the entryReference to video
AI video that keeps a character, product or place from your pictures recognizable in a new shot.
Read the entryTemporal consistency
How well an AI video keeps things the same from frame to frame, without flicker or drift.
Read the entryText to video
AI that turns a written prompt into a video clip, every frame invented from the words.
Read the entry
AI video editing
- Start here
AI video editing
AI video editing is using AI to change a video. Assistive tools speed up ordinary editing: cutting silences, adding captions, reframing. Generative editing, this page’s subject, changes what is in the frame: it swaps, removes, adds or restyles things in a real clip while its timing and camera stay as they were.
Read the entry Motion transfer
AI that takes the movement from one video and gives it to a different character.
Read the entryVideo inpainting
Filling a region of a video so it fits every frame, which is how AI removes objects and people.
Read the entry
Image editing
Image upscaling
Making a picture larger with more pixels, the extra detail predicted so it stays sharp.
Read the entryInpainting
Filling a chosen area of a picture with new content, the way AI removes objects and text.
Read the entryOutpainting
Extending a picture past its edges with new content that matches the scene.
Read the entry