Tool
Camera controls
A prompt says what happens; the camera says how it feels. eroq's composer carries 17 camera moves as one-click chips — from a locked-off static frame to a snorricam rig — plus lens character, aperture and lighting from the director's rack. They cumulate with your prompt server-side, in the register each engine understands.
50 free credits · no card · adults only, by policy
The moves
- →Everyday: Static, Slow pan, Push-in, Tracking, Orbit, Handheld, Close-up, Wide.
- →Cinephile rack: Pull-back, Dolly zoom, Crane up, Crane down, Whip pan, FPV drone, Aerial pull-back, POV, Snorricam.
- →Written directions work too — "wide shot" or "dolly in" inside the prompt is detected and applied.
Lens, aperture, light
Lenses: clean sharp, anamorphic, vintage anamorphic, soft portrait, macro, fisheye. Apertures: f/1.4 wide open, f/4, f/11 deep focus. Lighting: silhouette, practicals, window light, overhead fall, contre-jour, soft cross, neon glow. A soft-portrait lens at f/1.4 with a slow push-in is the intimate close-up; an anamorphic lens with a crane up and contre-jour is the trailer shot.
Pairing moves with engines
Slow moves (pan, push-in, orbit) render cleanly everywhere. Fast moves (whip pan, FPV, chaotic tempo) are best on Hailuo and Seedance. Handheld suits Kling's human motion and the UGC ad template. Keep one move per clip: stacking two in five seconds reads as a mistake.
Starting points
Every tile opens the studio with the prompt already written.
Frequently asked questions
Can I control the camera in AI video?
On eroq, yes: 17 named camera moves, six lenses, three apertures and seven lighting setups are one-click chips folded into the prompt, and written directions in the prompt are detected as well.
Do all engines understand camera moves?
All of them take the move as part of the prompt; the next-gen engines (Seedance, Kling, Hailuo, Veo) follow complex moves like dolly zoom and FPV most faithfully.
What is the difference between a push-in and a dolly zoom?
A push-in moves the camera toward the subject. A dolly zoom moves in while zooming out (or the reverse), so the subject stays the same size while the background stretches — the vertigo effect.
Keep exploring