MiniMax H3 Prompt:
A sensual anime fashion film follows a woman flowing through ten precise poses. @[char ref] defines her complete appearance, outfit, accessories, and anime rendering style; ignore the background and add no new accessories.
The film unfolds in one uninterrupted 15-second shot at 128 BPM inside a minimal white cyclorama with hard strobes, faint haze, glossy floor reflections, a satin sheet, and scattered proof sheets. The camera follows one physically continuous path, accelerating between poses and braking sharply on every lock.
0:00-0:01.5: Pose 1, centered frontal MCU. She faces forward with one hand brushing the hair beside her ear as the camera makes a restrained push-in.
0:01.5-0:03.0: Pose 2, she turns three-quarter and lifts her chin. Pose 3, she lowers her gaze as both hands settle softly near her collar and hair. The camera descends in a fast arc, then pauses on her eyes.
0:03.0-0:04.5: Pose 4, she rotates into a strict left profile. Pose 5, she rolls one shoulder forward while the robe edge catches a sharp side light. The camera whips past her cheek and settles close.
0:04.5-0:06.0: Pose 6, she gathers a fold of the robe at her waist. Pose 7, she releases it and turns her mouth toward the lens in an over-the-shoulder look. The camera dives to torso level and rises into a close facial pass.
0:06.0-0:07.5: Pose 8, she lowers onto the satin sheet with one knee raised. Pose 9, she extends one bare foot toward the lens, dominating the foreground. The camera rockets forward at floor level and briefly hangs on the foreshortened pose.
0:07.5-0:09.0: Pose 10, she rises into a three-quarter stance with one hand at her collarbone and the other touching her hair as the robe skims her thigh. The camera slides rapidly across the waistline, then eases into a portrait hold.
0:09-0:11.0: Without repeating a pose, she folds inward and closes her eyes for one breath, then opens into a long upward stretch. The camera circles in a tight orbit, slow on the locks and fast through the transition.
0:11.0-0:13.0: She twists into a rear three-quarter shoulder silhouette, then turns just enough for her jawline and robe neckline to catch the same flash. The camera skims continuously from shoulder to face with strong parallax.
0:13.0-0:15.0: The camera grazes past her eyes, follows the robe neckline, then arcs outward and lowers as she lands in a dominant full-body pose looking down into the lens. End on a clean wide hold under a final white flash bloom.
Use a smooth 24-85mm zoom feel, hard frontal flashes, sharp side rim light, shutter clicks, breath, hair movement, and fabric rustle. Keep one continuous character with stable identity, anatomy, hairstyle, outfit, accessories, and anime style throughout.
esse aqui é o tipo de site que eu queria ter achado antes de queimar crédito testando prompt no chute
46 movimentos de câmera com o vídeo do resultado e o prompt pronto pra copiar, em 7 categorias
de graça, salva
https://t.co/AtSCz6zzfi
How to make AI videos look more cinematic.
Add noise.
AI-generated videos can look too clean, making human skin look almost plastic. Adding noise as a layer of grain gives the footage a more film-like look and helps reduce the AI feel.
This technique is still widely used in films and music videos, even in today's digital-first world.
Some AI video creators may not know this, so I wanted to share it.
Edited with @capcutapp_jp. Watch in 4K.
Details below ↓
なんとなくアスペクト変更とアップスケールを内包してるプロンプト共有しときます。
プロンプト
Please convert the original 16:9 horizontal video @(specify video) into a native 9:16 vertical video, and simultaneously upscale it to a higher resolution.
Do not simply crop the original video; instead, intelligently reframe, extend, and recompose the scenes to create a vertical composition that looks as if it were originally produced in a 9:16 format.
Preserve all key visual information, characters, actions, objects, background details, and narrative progression from the original horizontal video. Naturally reposition elements within the vertical frame as needed to ensure no important elements are cut off.
Maintain complete consistency with the original video regarding characters, animation, action, timing, camera work, environment, lighting, color, saturation, and visual style.
Upscale resolution only. The upscaling must strictly preserve the original design: do not redesign, restyle, or reinterpret any element. Do not alter character designs, facial features, proportions, costumes, props, line work, shading style, textures, color palette, or art direction. Do not add, remove, or embellish any detail that does not exist in the original video. The upscale must only increase pixel resolution and sharpness, recovering fine detail that is already implied in the source, while eliminating compression artifacts, blur, noise, and aliasing.
Ensure correct anatomy, object shapes, perspective, spatial relationships, and overall coherence. Any areas newly generated outside the original frame must blend seamlessly with the existing environment and match the same resolution, sharpness, grain, and rendering quality as the upscaled original area.
Final output: 9:16 vertical format, high-resolution upscaled, full-frame, natural composition, no black bars, no image stretching, no loss of key details, no design changes, and visual consistency with the original 16:9 video.
One AI video skill. A full production team behind it. 🔥
이름은
[Luna AI Video Studio]
제가 AI 영상 만들 때마다 반복하던 일을 한 곳에 넣었습니다.
감독
카메라
배우 연기와 동선
대사
사운드
씬 연결
물리적인 움직임
레퍼런스 유지
AI 영상에서 자주 터지는 오류 체크까지.
사용법은 어렵게 만들고 싶지 않았어요.
그냥
“15초 좀비 공포영화 만들어줘.”
이렇게 시작해도 됩니다.
그러면 장면을 짜고,
카메라 위치와 움직임을 잡고,
배우가 언제 움직이고 어디를 볼지 정하고,
대사와 ��운드가 필요한지도 판단해서
영상 생성기에 바로 넣을 수 있는 프롬프트까지 만들어줘요.
조금 더 원하는 게 있다면 이렇게 말하면 되고요.
“이 이미지로 10초 향수 광고 만들어줘. Seedance 2.5로.”
“앞 장면 이어서 8초 더 만들어줘.”
“완성본 봐주고 이상한 부분 수정 프롬프트 만들어줘.”
제가 AI 영상 만들면서 제일 스트레스 받았던 것도 넣었습니다.
씬이 바뀔 때마다 얼굴이 바뀌고,
소품이 사라지고,
사람은 걷는데 배경은 멈춰 있고,
카메라는 갑자기 엉뚱한 곳으로 가고...
이런 문제를 생성 전에 잡도록 해뒀어요.
그리고 이번에는 여기서 한 단계 더 갔습니다.
생성 결과가 이상하면
어디서 실패했는지 찾고
→ 필요한 부분만 수정하고
→ 다시 생성할 수 있게 했어요.
긴 영상에서는 이전 씬의 얼굴, 의상, 소품, 이동 방향, 상태도 다음 씬으로 이어가게 만들었습니다.
Kling, Seedance, Veo 같은 영상 모델도 지정할 수 있고
버전까지 알려주면 그 모델에 맞게 프롬프트를 조정합니다.
Claude Code, Codex, Gemini CLI처럼
Agent Skills를 지���하는 환경에서 사용할 수 있어요.
제가 만들고 싶었던 건 딱 하나였습니다.
아이디어만 있는 사람도
“이런 영상 만들고 싶어.”
이 한마디로 시작할 수 있는 AI 영상 제작팀.
스킬 주소는 댓글에 둘게요.
써보고 영상 나오면 꼭 보여주세요.
저도 진짜 어디까지 만들어지는지 궁금합니다.
(아래 영상은 루나 캐릭터시트 단 한장만 넣고, “15초 좀비 공포영화 만들어줘.” 단 한줄로 스킬에서 자동생성해 준 프롬프트로 생성된 첫번째 영상입니다.)