Art style: Cinematic Realistic. Char_0 standing in the near foreground; char_1 standing in the middle ground. Very close proximity, almost touching. Char_0 is standing on pavement in the near foreground in the midground. Torso in three-quarter turn to screen-right. Head in three-quarter turn to screen-right. Eyes on char_1. Holding a paper cup. Shocked. Char_0 wears fixed_by_reference. Partially behind none. Char_1 is standing on pavement in the middle ground in the midground. Torso in three-quarter turn to screen-left. Head in three-quarter turn to screen-left. Eyes on char_0. At sides. Annoyed. Char_1 wears fixed_by_reference. Partially behind none. Wide shot from a low angle, 9:16. Horizon middle. Dynamic collision focus. Foreground: pavement with coffee splashes. Midground: characters and street furniture. Background: pink tuff buildings of Yerevan. Key light front_left, bright sunlight. Sharp. Focus: characters. Dominant palette: terracotta, beige, warm gold, coffee brown. SPATIAL LOCK: very close proximity, almost touching. GAZE LOCK: char_0 head in three-quarter turn to screen-right eyes on char_1; char_1 head in three-quarter turn to screen-left eyes on char_0. This is a single frozen instant (action start) — no motion blur, no action chained. ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ FINAL STYLE ENFORCEMENT (this is the authoritative rendering directive — it overrides any style wording elsewhere in the prompt): Render this image entirely in "Cinematic Realistic" style. Match the lighting grammar, material response, colour science, level of realism / stylisation, line weight, and shading model typical of "Cinematic Realistic". Keep the subject, composition, camera angle, lighting direction, and colour palette from the description above — but EVERY surface, skin, fabric, material, and environmental element must be rendered in the "Cinematic Realistic" idiom, not in any other medium.
Free to start · Generate videos and images with AI in seconds