e2b964ca-6496-4212-9703-b9669428cf91
- Queued
- Jul 20, 2026, 06:32 PM
- Completed
- Jul 20, 2026, 06:32 PM
- Execution time
- 11.4s
- Outputs
- 1 · 1.1 MB
- Inputs
- 1
- Cached nodes
- 1
Outputs
1Click an image to view it full size.
Text prompts
3Extreme close-up of @Elena's grey-green eyes reflected in a car rearview mirror. Only her eyes and the bridge of her nose are visible in the mirror — the rest of her face is cut by the mirror's edge. Her eyes are looking forward through the windshield, not at herself. They carry a weight that does not belong to a twenty-eight-year-old — heavy upper lids, faint shadows beneath, watchful and exhausted and ancient. The mirror frame is slightly dirty. Through the windshield behind her reflection: the blurred grey shapes of coastal houses appearing through fog. The mirror is positioned in the upper-centre of the vertical frame. Below it: the soft blur of her hands on the steering wheel, the dashboard, the old photograph face-down on the passenger seat. Shot on full-frame mirrorless, 85mm at f1.8 — extremely shallow, her eyes in the mirror critically sharp, everything else soft. Natural flat light from the windshield illuminating her face evenly, no direct sun, no colour grade. Hyperrealistic — the fine blood vessels in her sclera, individual lashes, the specific grey-green of her irises, dust on the mirror glass, the worn leather of the steering wheel.
Generate a single, hyper-realistic character reference image (strictly no collages, text, watermarks, or borders). The image must feature a cinematic, three-quarter portrait of the described character, maintaining a sharp, tack-focused emphasis on the face with natural skin textures, pore details, and realistic micro-expressions. The lighting should be cinematic and directional (e.g., subtle split or Rembrandt lighting) to give depth, blending seamlessly with the environment specified in the character description. The character must have a natural, living presence, looking slightly off-camera or directly at the lens with a subtle, context-aware expression. Ensure the full head, neck, and upper shoulders are completely unobstructed. This image serves as a high-fidelity identity blueprint for temporal-consistent video generation.
1 other text field— sigmas, filenames, and other non-prompt strings the logger swept up
node 4 · LoadImage · image
WhatsApp Image 2026-07-21 at 12.02.04 AM.jpeg
Input media
1Source images and video this run consumed.