Advanced Prompting
Advanced prompt engineering techniques for professional-grade AI video generation
Prompt Structure
A well-structured VisioArt prompt is built from layered parameters. Each parameter adds specificity that guides the model toward your intended output.
| Parameter | Purpose | Example |
|---|---|---|
| Subject | Who or what is in the scene | "a woman in a red dress" |
| Motion | How the subject moves | "walking slowly through a crowd" |
| Environment | Where the scene takes place | "rainy Tokyo street at night" |
| Lighting | Quality and source of light | "neon reflections on wet pavement" |
| Style | Overall visual treatment | "cinematic, film noir" |
| Camera | How the shot is framed or moves | "slow tracking shot" |
| Mood | Emotional tone | "melancholic, tense" |
Combine all parameters in a single prompt sentence for best results: "A woman in a red dress walking slowly through a rainy Tokyo crowd at night, neon reflections on wet pavement, cinematic film noir, slow tracking shot, melancholic mood."
Style Modifier Keywords
Add these to your prompt to steer the visual aesthetic:
cinematic— widescreen composition with dramatic depth of fieldhyperrealistic— photographic detail and natural imperfectionsslow motion— stretched time with motion blurdrone shot— elevated aerial perspectivebokeh— sharp subject against a softly blurred backgroundgolden hour— warm, low-angle sunlight at dusk or dawnfilm noir— high contrast black and white with dramatic shadowsanime style— Japanese animation cel-shading and proportions
Camera Movement Keywords
Specify how the virtual camera behaves during the shot:
tracking shot— camera follows the subject laterallyzoom in— gradual magnification toward the subjectpan left / pan right— horizontal pivot from a fixed positionaerial view— top-down perspective from above the scenedolly shot— camera physically moves forward or backward on a trackstatic shot— locked-off camera with no movement
Lighting Keywords
Lighting has one of the largest impacts on perceived quality:
soft diffused— overcast or studio box-light quality, no harsh shadowsdramatic backlighting— subject silhouetted against a bright backgroundneon glow— colored artificial light spilling onto surfacescandlelight— warm, flickering low-key illuminationstudio lighting— clean three-point lighting setup for portrait clarity
Motion Descriptors
Guide how elements within the scene move:
gentle waves— slow, rhythmic water or fabric motionswaying trees— natural wind-driven plant movementflowing hair— loose hair reacting to movement or breezebouncing walk— energetic, upbeat pedestrian motionexplosive action— fast, high-energy burst movement
Troubleshooting Prompt Mismatches
If your output does not match your intent, the most common cause is contradictory or ambiguous instructions.
- Too vague: "A nice video of nature" gives the model no constraints. Add subject, location, lighting, and style.
- Contradictory motion: Do not combine
slow motionwithexplosive action— choose one. - Style conflicts: Mixing
anime stylewithhyperrealisticproduces inconsistent output. Pick a single dominant style. - Over-specified prompts: More than 200 words can cause models to drop details. Keep prompts focused and under 150 words.
- Wrong model for the style: Anime-style prompts perform best on Z-Image or Seedance, not on photorealistic models like Veo 3.1.
VisioArt Docs