[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"sales-offer:seasonal-sale:public":-1,"blog:article:movimentos-de-camera-para-ia":3},{"article":4},{"id":5,"slug":6,"title":7,"excerpt":8,"category":9,"tags":10,"cover":18,"author":22,"readingMinutes":26,"publishedAt":27,"featured":28,"blocks":29,"toc":577,"modelKeys":613,"skills":614,"updatedAt":615,"seo":616,"related":619,"commentCount":670},"6af3ac113383dc208a10061a","movimentos-de-camera-para-ia","Movimentos de câmera para IA: 20 comandos de cinema no Seedance 2.5 Edit + Claude (com os prompts)","Gravei um take só, com a câmera parada, e pedi o mesmo vídeo com mais de 20 movimentos de cinema no Seedance 2.5 Edit da KAIROGEN. Aqui estão os nomes em português e em inglês, para que serve cada um e o prompt pronto para colar.","guias",[11,12,13,14,15,16,17],"movimentos de câmera","movimento de câmera ia","seedance 2.5","prompt","claude","video com ia","kairogen",{"kind":19,"url":20,"alt":21},"image","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa3524e81-c7af-4fb5-9189-5311a36dbb6d.jpg","Movimentos de câmera para IA: Gabriel apontando para três quadros do mesmo take refilmado por IA, um FPV sobre a piscina, uma plongée de cima e uma câmera de baixo. Texto: 20 MOVIMENTOS DE CÂMERA",{"name":23,"handle":24,"avatarUrl":25},"Gabriel Adamuchi","adamuchi.gabriel","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F15a97e0a-60dc-40d4-9aaf-0147ee5260e5.jpg",10,"2026-11-11T00:06:33.197Z",false,[30,34,54,60,63,67,70,76,79,84,87,99,108,112,115,123,128,132,135,143,148,152,155,163,168,172,175,179,182,190,195,199,202,210,215,219,222,230,235,239,242,250,254,257,265,270,273,277,280,288,293,297,300,308,313,317,320,328,333,337,340,348,353,356,360,363,371,376,380,383,391,396,400,403,411,416,420,423,431,436,439,443,446,454,459,463,466,474,479,483,486,494,499,503,506,514,519,522,525,531,534,537,544,550,553,556,559,562,565,568,571,574],{"id":31,"type":32,"markdown":33},"abre","text","Os movimentos de câmera para IA têm nome, e esse nome é o que faz o vídeo parecer filme. Eu gravei um take só, no quintal, com o celular parado num tripé, e pedi ao Seedance 2.5 Edit, na KAIROGEN, o mesmo take filmado por outra câmera: aproximação lenta, órbita, grua, drone, FPV, chicote e mais de 20 movimentos.\n\nA minha fala, a minha boca e as minhas mãos continuam no mesmo segundo do original. Muda só a câmera. Abaixo estão todos os comandos, com o nome em português, o termo de cinema em inglês, para que cada um serve e o prompt pronto para colar.",{"id":35,"type":36,"title":37,"items":38},"como","steps","Como funciona",[39,42,45,48,51],{"title":40,"body":41},"Grave um take só, com a câmera parada","Plano aberto, de corpo inteiro, com folga em volta. Quanto mais espaço no quadro, mais a câmera de IA tem para andar.",{"title":43,"body":44},"Corte um trecho por movimento","O modo edit do Seedance 2.5 aceita vídeo de 4 a 30 segundos. Eu cortei trechos de 4 segundos, um para cada movimento.",{"title":46,"body":47},"Peça um movimento por geração","Vários movimentos num pedido só o modelo junta ou atrasa. Um movimento por trecho sai certo, e quem coloca cada um no segundo da fala é a edição.",{"title":49,"body":50},"Faça o rascunho em 480p","Aprove a câmera no rascunho, que sai barato. Depois refaça em 1080p usando o próprio rascunho como vídeo de origem.",{"title":52,"body":53},"Mande pelo Claude ou pelo estúdio","Com o Claude conectado na KAIROGEN pelo MCP, ele corta os trechos, manda as gerações e junta tudo. Também dá para subir cada trecho no estúdio de vídeo e colar o prompt.",{"id":55,"type":56,"text":57,"level":58,"eyebrow":59},"lista-h","heading","Movimentos de câmera: os nomes em português e em inglês",2,"Lista",{"id":61,"type":32,"markdown":62},"lista-t","O nome em português é como eu falo no vídeo. O termo em inglês, entre crases, é o que vai no prompt.\n\nPara apresentar alguém com peso\n\n- Aproximação lenta: `slow push-in`\n- Câmera de baixo: `low angle`\n- Traveling lateral: `lateral tracking shot`\n- Órbita: `orbit`\n\nPara dar tensão\n\n- Dolly in: `dolly in`\n- Zoom: `snap zoom`\n- Câmera na mão: `handheld camera`\n- Plano holandês: `dutch angle`\n- Close extremo: `extreme close-up`\n\nPara revelar um lugar\n\n- Grua subindo: `crane up`\n- Drone se afastando: `drone pulling away`\n- Câmera de cinema: `cinematic wide glide`\n- Afastamento: `pull back`\n\nPara dar energia\n\n- Chicote: `whip pan`\n- FPV: `FPV drone`\n- Câmera presa: `body-mounted camera` (snorricam)\n- Câmera girando: `spinning camera` (barrel roll)\n\nExtras do vídeo\n\n- Dolly-zoom: `dolly zoom` (vertigo)\n- Plongée de cima: `top-down crane down`\n- Close nas mãos: `close-up on the hands`\n- Close na boca: `extreme close-up on the mouth`",{"id":64,"type":56,"text":65,"level":58,"eyebrow":66},"trava-h","Como escrever o prompt de movimento de câmera","Método",{"id":68,"type":32,"markdown":69},"trava-t","Todo prompt tem três partes. Primeiro a trava de performance: é o mesmo take, só a câmera muda, cada gesto no mesmo segundo e a boca no tempo do áudio original. Depois o movimento, com o nome em inglês e uma frase sobre como a câmera anda. Por último, o que não pode aparecer.\n\nOs prompts abaixo já vêm com as três partes. Para usar no seu vídeo, suba o seu trecho como vídeo de origem no Seedance 2.5 Edit e cole o prompt do movimento que você quer.",{"id":71,"type":72,"tone":73,"title":74,"markdown":75},"dica-tempo","callout","tip","Comece o trecho 2,5 segundos antes da fala","Mudança de enquadramento, como close ou câmera de baixo, já entra no primeiro quadro. Movimento de câmera, como grua, órbita e drone, leva uns 2,5 segundos para ganhar força. Então corte o trecho de movimento começando 2,5 segundos antes da palavra, e use na edição a parte em que a câmera já está andando.",{"id":77,"type":56,"text":78,"level":58},"g-peso","Movimentos de câmera no cinema para apresentar alguém com peso",{"id":80,"type":56,"text":81,"level":82,"eyebrow":83},"m01-h","Aproximação lenta (slow push-in)",3,"01",{"id":85,"type":32,"markdown":86},"m01-t","A câmera vem andando devagar até você, do corpo inteiro ao plano médio. É o movimento clássico de quem entra em cena com importância.",{"id":88,"type":89,"before":90,"after":94,"beforeLabel":97,"afterLabel":98},"m01-c","compare",{"kind":91,"url":92,"posterUrl":93},"video","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F7b974a49-80cf-4a32-8115-ebdea8a57230.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F8ece03bb-33fe-44ba-9d13-63bf78218d32.jpg",{"kind":91,"url":95,"posterUrl":96},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F61b43218-2a8f-4a7b-abd7-65cf33444f43.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F6c449922-1590-4064-b497-03291024758d.jpg","Original","Com IA",{"id":100,"type":14,"title":101,"studio":91,"modelKey":102,"modelName":103,"prompt":104,"params":105,"resultUrl":95,"resultKind":91},"m01-p","Prompt: aproximação lenta (slow push-in)","seedance-2-5","Seedance 2.5 Edit","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. SLOW PUSH-IN: the camera travels slowly and steadily forward toward the person, from a wide full-body shot to a medium shot, frontal, eye level, smooth, giving weight and importance. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},"9:16","480p",{"id":109,"type":56,"text":110,"level":82,"eyebrow":111},"m02-h","Câmera de baixo (low angle)","02",{"id":113,"type":32,"markdown":114},"m02-t","A câmera fica perto do chão olhando para cima, com o céu atrás da cabeça. Passa poder e confiança.",{"id":116,"type":89,"before":117,"after":120,"beforeLabel":97,"afterLabel":98},"m02-c",{"kind":91,"url":118,"posterUrl":119},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F7c6bb990-7f1b-4a6a-8557-2ed37c399c91.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F3c54134c-7df7-4ede-bd20-7f4b81e3b4a8.jpg",{"kind":91,"url":121,"posterUrl":122},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F673788f6-a04f-4390-b0fd-d405e39f2819.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F8316f2e6-6132-4817-913b-7190157518e4.jpg",{"id":124,"type":14,"title":125,"studio":91,"modelKey":102,"modelName":103,"prompt":126,"params":127,"resultUrl":121,"resultKind":91},"m02-p","Prompt: câmera de baixo (low angle)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single framing, from the very first frame and held to the end): LOW ANGLE: the camera is close to the ground below the person, looking up, sky behind the head, heroic and powerful, a slow gentle push. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":129,"type":56,"text":130,"level":82,"eyebrow":131},"m03-h","Traveling lateral (lateral tracking shot)","03",{"id":133,"type":32,"markdown":134},"m03-t","A câmera desliza de lado, em paralelo a você, e o fundo passa por trás. Dá ritmo sem tirar o foco de quem fala.",{"id":136,"type":89,"before":137,"after":140,"beforeLabel":97,"afterLabel":98},"m03-c",{"kind":91,"url":138,"posterUrl":139},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fee421536-9dca-4431-bf48-c5fa7f081b96.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fb3d75f96-03d1-40cd-92c6-0808dc2d8a23.jpg",{"kind":91,"url":141,"posterUrl":142},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa41c56b4-7974-4097-bfe2-7be46cbe3db3.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fd74426a0-ea0a-48fc-a503-3272abcd2a78.jpg",{"id":144,"type":14,"title":145,"studio":91,"modelKey":102,"modelName":103,"prompt":146,"params":147,"resultUrl":141,"resultKind":91},"m03-p","Prompt: traveling lateral (lateral tracking shot)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. LATERAL TRACKING SHOT: the camera slides sideways from left to right, parallel to the person, at medium distance, the background sliding past, smooth dolly track feel. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":149,"type":56,"text":150,"level":82,"eyebrow":151},"m04-h","Órbita (orbit)","04",{"id":153,"type":32,"markdown":154},"m04-t","A câmera dá a volta em você pela frente, com o rosto sempre visível. O fundo gira e o plano ganha profundidade.",{"id":156,"type":89,"before":157,"after":160,"beforeLabel":97,"afterLabel":98},"m04-c",{"kind":91,"url":158,"posterUrl":159},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F78f2b4d3-0cae-479e-8256-a275c1a43509.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Ffc229bfa-a426-484f-b10a-71c960a3c30c.jpg",{"kind":91,"url":161,"posterUrl":162},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F975cc973-5cad-46c7-8ef2-9ffaca52c425.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa62f13c1-8d2e-4c1e-8485-6eecbc15833f.jpg",{"id":164,"type":14,"title":165,"studio":91,"modelKey":102,"modelName":103,"prompt":166,"params":167,"resultUrl":161,"resultKind":91},"m04-p","Prompt: órbita (orbit)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. ORBIT: the camera circles around the person at medium distance at a steady pace, from the right side across the front toward the left side, always keeping the face visible, strong parallax in the background. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":169,"type":72,"tone":73,"title":170,"markdown":171},"dica-rosto","Em close, mande a foto do seu rosto junto","Quando o movimento chega perto, como no dolly in, no zoom e nos closes, suba duas ou três fotos nítidas do seu rosto como imagem de referência, de frente e de lado. O modelo usa as fotos para manter o seu rosto igual no plano fechado.",{"id":173,"type":56,"text":174,"level":58},"g-tensao","Tipos de movimento de câmera para dar tensão",{"id":176,"type":56,"text":177,"level":82,"eyebrow":178},"m05-h","Dolly in (dolly in)","05",{"id":180,"type":32,"markdown":181},"m05-t","A câmera avança rápido até o rosto, do plano médio ao close. É o movimento de suspense, de quem vai contar um segredo.",{"id":183,"type":89,"before":184,"after":187,"beforeLabel":97,"afterLabel":98},"m05-c",{"kind":91,"url":185,"posterUrl":186},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F2b39728d-0728-49ea-917c-73d89032f23a.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F3a527edf-b4df-47cd-bcb5-ea78fb528c70.jpg",{"kind":91,"url":188,"posterUrl":189},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa0314a8b-b781-48da-bb02-7b2edf7f504b.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fc4f378d4-ed44-4593-8e98-fc49fcd7c8e7.jpg",{"id":191,"type":14,"title":192,"studio":91,"modelKey":102,"modelName":103,"prompt":193,"params":194,"resultUrl":188,"resultKind":91},"m05-p","Prompt: dolly in (dolly in)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. DOLLY IN: the camera physically travels forward toward the face, fast and tense, from a medium shot to a tight shot of the face and shoulders, thriller mood. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":196,"type":56,"text":197,"level":82,"eyebrow":198},"m06-h","Zoom (snap zoom)","06",{"id":200,"type":32,"markdown":201},"m06-t","Um zoom seco e muito rápido do plano aberto até o rosto. Funciona como uma pancada na hora da frase forte.",{"id":203,"type":89,"before":204,"after":207,"beforeLabel":97,"afterLabel":98},"m06-c",{"kind":91,"url":205,"posterUrl":206},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F45a85c3a-4bc4-4a1e-966e-694ca2754216.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F7b5d3a86-f3eb-48ce-8e9d-6bb6ed247c0f.jpg",{"kind":91,"url":208,"posterUrl":209},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fb6b61db9-54d7-4f51-9b49-3edb7f843010.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F97730a48-08ac-45c8-9196-ab51b2fa32e7.jpg",{"id":211,"type":14,"title":212,"studio":91,"modelKey":102,"modelName":103,"prompt":213,"params":214,"resultUrl":208,"resultKind":91},"m06-p","Prompt: zoom (snap zoom)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. SNAP ZOOM: a sudden, very fast optical zoom-in from a wide shot to a tight shot of the face, the background compressing, then it holds tight. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":216,"type":56,"text":217,"level":82,"eyebrow":218},"m07b-h","Câmera na mão (handheld camera)","07",{"id":220,"type":32,"markdown":221},"m07b-t","A câmera treme e se ajusta como se alguém estivesse segurando, no estilo documentário. Deixa o vídeo tenso e real.",{"id":223,"type":89,"before":224,"after":227,"beforeLabel":97,"afterLabel":98},"m07b-c",{"kind":91,"url":225,"posterUrl":226},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fbb7be55b-f622-4ff9-9a8a-a2ec5097be0e.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fafba612e-d7dc-4fd6-a5bf-048cd46cd5b4.jpg",{"kind":91,"url":228,"posterUrl":229},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F37abbac7-89ad-4433-85e7-93eee1b5d443.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fba099221-bd75-453c-b7ee-076d478d2302.jpg",{"id":231,"type":14,"title":232,"studio":91,"modelKey":102,"modelName":103,"prompt":233,"params":234,"resultUrl":228,"resultKind":91},"m07b-p","Prompt: câmera na mão (handheld camera)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. HANDHELD CAMERA: nervous handheld camera operated by hand, visible shake, small drifts and reframes, documentary style, medium close shot, tense. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":236,"type":56,"text":237,"level":82,"eyebrow":238},"m08-h","Plano holandês (dutch angle)","08",{"id":240,"type":32,"markdown":241},"m08-t","A câmera fica inclinada e o horizonte vira uma diagonal. Dá a sensação de que algo está fora do lugar, como em filme de suspense.\n\nO jeito mais preciso de fazer o holandês é na edição: gere o plano médio parado e, no editor, gire o quadro 20 graus e amplie cerca de 145%, até a borda sumir. Assim você escolhe a inclinação exata e ela fica firme do começo ao fim.",{"id":243,"type":89,"before":244,"after":247,"beforeLabel":97,"afterLabel":98},"m08-c",{"kind":91,"url":245,"posterUrl":246},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F92ed201f-06eb-4191-a0d5-20fe7af0cd19.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fb53bbac5-8dc9-4fa8-b427-6ea24e36df90.jpg",{"kind":91,"url":248,"posterUrl":249},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F93ec0273-eff6-4dac-91ff-9ac6e65f1092.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F77f822c0-a1bb-4659-a0f5-9fad17229525.jpg",{"id":251,"type":56,"text":252,"level":82,"eyebrow":253},"m09-h","Close extremo (extreme close-up)","09",{"id":255,"type":32,"markdown":256},"m09-t","Do supercílio ao queixo, com os olhos e a boca enchendo o quadro. É o close do momento mais sério do vídeo.",{"id":258,"type":89,"before":259,"after":262,"beforeLabel":97,"afterLabel":98},"m09-c",{"kind":91,"url":260,"posterUrl":261},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F0f7122b9-5aee-46c4-9d8d-a1ff95bd73fa.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F68e9d74e-eab1-4545-8a31-12c3b1c32e34.jpg",{"kind":91,"url":263,"posterUrl":264},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa3493bfa-d305-4d85-8b4f-6f48f806270b.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fd88e5161-be1f-4957-acc5-454f49fc926a.jpg",{"id":266,"type":14,"title":267,"studio":91,"modelKey":102,"modelName":103,"prompt":268,"params":269,"resultUrl":263,"resultKind":91},"m09-p","Prompt: close extremo (extreme close-up)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move, it starts on the very first frame): EXTREME CLOSE-UP. From frame 1 the frame goes from the eyebrows to the chin, frontal, eye level, level horizon. The eyes and mouth fill the frame, sharp focus on the eyes, the background blurred. The camera holds this extreme close-up for the whole clip with an almost imperceptible push-in; it never pulls back and never orbits to the side.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":271,"type":56,"text":272,"level":58},"g-lugar","Movimentos de câmera para revelar um lugar",{"id":274,"type":56,"text":275,"level":82,"eyebrow":276},"m10-h","Grua subindo (crane up)","10",{"id":278,"type":32,"markdown":279},"m10-t","A câmera sobe alto como uma grua e olha para baixo, mostrando o lugar inteiro em volta de você.",{"id":281,"type":89,"before":282,"after":285,"beforeLabel":97,"afterLabel":98},"m10-c",{"kind":91,"url":283,"posterUrl":284},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F22d6ded4-9d72-464d-968c-45fef95009d7.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F54a9f8fd-d181-4ee9-be79-2ca9861b90ae.jpg",{"kind":91,"url":286,"posterUrl":287},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F5157108b-68f7-4f21-a39d-895bc3b407bd.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa3072a39-eab5-4197-b94c-26cf8da5e95c.jpg",{"id":289,"type":14,"title":290,"studio":91,"modelKey":102,"modelName":103,"prompt":291,"params":292,"resultUrl":286,"resultKind":91},"m10-p","Prompt: grua subindo (crane up)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. CRANE UP: the camera rises high above the person like a crane, tilting down, revealing the whole place around. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":294,"type":56,"text":295,"level":82,"eyebrow":296},"m11-h","Drone se afastando (drone pulling away)","11",{"id":298,"type":32,"markdown":299},"m11-t","A câmera voa para trás e para cima, como um drone, e revela o lugar e o que está além dele.",{"id":301,"type":89,"before":302,"after":305,"beforeLabel":97,"afterLabel":98},"m11-c",{"kind":91,"url":303,"posterUrl":304},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fe1cc26af-4c2a-4b60-83bf-9ee3f9277810.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fff16e092-4355-48ce-adc7-18af4611bc09.jpg",{"kind":91,"url":306,"posterUrl":307},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F1bc92d51-cd15-480e-91dc-709b239a7236.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F2d72c3eb-93ff-45ab-80b2-0deb8a2fdce7.jpg",{"id":309,"type":14,"title":310,"studio":91,"modelKey":102,"modelName":103,"prompt":311,"params":312,"resultUrl":306,"resultKind":91},"m11-p","Prompt: drone se afastando (drone pulling away)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. DRONE PULLING AWAY: the camera flies backward and up like a drone, revealing the whole location and what is beyond it; the person stays clearly visible in the center. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":314,"type":56,"text":315,"level":82,"eyebrow":316},"m12-h","Câmera de cinema (cinematic wide glide)","12",{"id":318,"type":32,"markdown":319},"m12-t","Um plano aberto e baixo, deslizando de lado, com lente grande angular e fundo desfocado. É a cara de cena de filme.",{"id":321,"type":89,"before":322,"after":325,"beforeLabel":97,"afterLabel":98},"m12-c",{"kind":91,"url":323,"posterUrl":324},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F4a017cad-a8be-4147-8d07-0acbfda177bb.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F2a29e603-e381-401c-a941-2a2cd22992b6.jpg",{"kind":91,"url":326,"posterUrl":327},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fe61fe188-837d-4287-b847-baaf90c2de77.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F583f9abd-8362-4778-ae6c-c635d270b4de.jpg",{"id":329,"type":14,"title":330,"studio":91,"modelKey":102,"modelName":103,"prompt":331,"params":332,"resultUrl":326,"resultKind":91},"m12-p","Prompt: câmera de cinema (cinematic wide glide)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. CINEMATIC CAMERA: a low wide cinematic shot gliding slowly sideways, wide lens, shallow depth of field, rich film look, like a scene from a movie. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":334,"type":56,"text":335,"level":82,"eyebrow":336},"m13-h","Afastamento (pull back)","13",{"id":338,"type":32,"markdown":339},"m13-t","A câmera recua em linha reta, do plano médio ao corpo inteiro. Bom para fechar uma ideia e abrir a próxima.",{"id":341,"type":89,"before":342,"after":345,"beforeLabel":97,"afterLabel":98},"m13-c",{"kind":91,"url":343,"posterUrl":344},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F0b06f1f5-c65a-47d7-967f-f69a83582330.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa6fe535b-f9bc-459e-9f33-906b775600dd.jpg",{"kind":91,"url":346,"posterUrl":347},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F1ec80ce7-cc32-48b9-9546-326aee7fff16.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F90106fb3-4b82-42aa-9078-dee23af8cf67.jpg",{"id":349,"type":14,"title":350,"studio":91,"modelKey":102,"modelName":103,"prompt":351,"params":352,"resultUrl":346,"resultKind":91},"m13-p","Prompt: afastamento (pull back)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. PULL BACK: the camera dollies straight backward away from the person, from a medium shot to a wide full-body shot, frontal, eye level, smooth. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":354,"type":56,"text":355,"level":58},"g-energia","Movimentos de câmera para dar energia",{"id":357,"type":56,"text":358,"level":82,"eyebrow":359},"m14-h","Chicote (whip pan)","14",{"id":361,"type":32,"markdown":362},"m14-t","A câmera vira de lado muito rápido, com borrão de movimento, e cai em você de outro ângulo. É a transição mais usada em vídeo curto.",{"id":364,"type":89,"before":365,"after":368,"beforeLabel":97,"afterLabel":98},"m14-c",{"kind":91,"url":366,"posterUrl":367},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F0ea26846-3b4f-4be2-900b-9a5653c6ce96.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F513c872d-06f5-4b10-b0f1-d2447bf19748.jpg",{"kind":91,"url":369,"posterUrl":370},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F9a6be6d9-4525-4ad6-8020-212c19f3c7f8.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F1474d9c0-1c47-4d91-8fde-1475505fc7de.jpg",{"id":372,"type":14,"title":373,"studio":91,"modelKey":102,"modelName":103,"prompt":374,"params":375,"resultUrl":369,"resultKind":91},"m14-p","Prompt: chicote (whip pan)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. WHIP PAN: the camera whips sideways extremely fast with heavy motion blur and lands on the person from a new angle, high energy, then whips again. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":377,"type":56,"text":378,"level":82,"eyebrow":379},"m15-h","FPV (FPV drone)","15",{"id":381,"type":32,"markdown":382},"m15-t","Um drone de corrida desce do alto, passa rente ao chão e vem na sua direção fazendo curva. É o movimento que mais prende o olho.",{"id":384,"type":89,"before":385,"after":388,"beforeLabel":97,"afterLabel":98},"m15-c",{"kind":91,"url":386,"posterUrl":387},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F2155ec83-e323-4d2f-8e9d-b33b754bae31.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fc2701951-44a1-4cee-8b2e-cfca15074e79.jpg",{"kind":91,"url":389,"posterUrl":390},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fed38e34d-f536-44ad-a01f-34204dc0aafb.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F1e81f501-02a7-4550-aee9-d0ef3afb94b9.jpg",{"id":392,"type":14,"title":393,"studio":91,"modelKey":102,"modelName":103,"prompt":394,"params":395,"resultUrl":389,"resultKind":91},"m15-p","Prompt: fpv (FPV drone)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. FPV DRONE: a fast FPV drone flies in from high above, dives low over the ground and races toward the person, banking, high energy. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":397,"type":56,"text":398,"level":82,"eyebrow":399},"m16-h","Câmera presa (body-mounted camera)","16",{"id":401,"type":32,"markdown":402},"m16-t","A câmera fica presa no seu corpo, apontada para o rosto: você fica parado no quadro e o lugar balança atrás. É o snorricam dos filmes.",{"id":404,"type":89,"before":405,"after":408,"beforeLabel":97,"afterLabel":98},"m16-c",{"kind":91,"url":406,"posterUrl":407},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fc6ea170a-3534-4b06-9aa6-4060f45190d1.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fc7f51f3f-8c7a-44cd-b3cf-a29c176818d4.jpg",{"kind":91,"url":409,"posterUrl":410},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F2c2bed1b-1fb4-45d6-b150-0eaf2733ee9d.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Ff932ac2d-63cd-4c48-850a-4009e878652f.jpg",{"id":412,"type":14,"title":413,"studio":91,"modelKey":102,"modelName":103,"prompt":414,"params":415,"resultUrl":409,"resultKind":91},"m16-p","Prompt: câmera presa (body-mounted camera)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. BODY-MOUNTED CAMERA: the camera is rigidly locked to the torso facing the face, snorricam style, the face and chest stay fixed in the frame while the location sways and moves behind. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":417,"type":56,"text":418,"level":82,"eyebrow":419},"m17-h","Câmera girando (spinning camera)","17",{"id":421,"type":32,"markdown":422},"m17-t","A câmera gira em volta de você e no próprio eixo, com o horizonte rodando. Bom para abrir o vídeo com energia.",{"id":424,"type":89,"before":425,"after":428,"beforeLabel":97,"afterLabel":98},"m17-c",{"kind":91,"url":426,"posterUrl":427},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F76887da3-deca-4466-8a34-95ab3b35791e.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F7fefba73-34ee-45e8-8127-d9ca20fc8363.jpg",{"kind":91,"url":429,"posterUrl":430},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F5faf4a44-9611-4ea1-a550-d95d1e190055.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fcdc623e4-cca4-47e1-bb43-8727e9dba42c.jpg",{"id":432,"type":14,"title":433,"studio":91,"modelKey":102,"modelName":103,"prompt":434,"params":435,"resultUrl":429,"resultKind":91},"m17-p","Prompt: câmera girando (spinning camera)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): the camera starts moving on the very first frame and the move must be fully established and clearly visible by 2.5s, then it keeps going strongly until the end of the clip. SPINNING CAMERA: the camera circles the person while rolling around its own axis, the horizon rotating continuously, high energy. The person stands still on the mark and does not walk.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":437,"type":56,"text":438,"level":58},"g-extras","Mais movimentos de câmera usados no vídeo",{"id":440,"type":56,"text":441,"level":82,"eyebrow":442},"b1-h","Dolly-zoom (dolly zoom (vertigo))","18",{"id":444,"type":32,"markdown":445},"b1-t","A câmera recua junto com você enquanto o fundo estica. É o efeito vertigo, que eu usei no gancho do vídeo, com chicote e grua logo depois.",{"id":447,"type":89,"before":448,"after":451,"beforeLabel":97,"afterLabel":98},"b1-c",{"kind":91,"url":449,"posterUrl":450},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fb8d6d4bc-cf54-4912-8fd8-9e1833832945.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fdb5cd7fe-165c-4792-af27-7b7603a78a6b.jpg",{"kind":91,"url":452,"posterUrl":453},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fc6a87d40-d8a1-4a1b-bc45-dadce7b22604.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F3e17d673-6c26-438a-b99a-9b380a96126f.jpg",{"id":455,"type":14,"title":456,"studio":91,"modelKey":102,"modelName":103,"prompt":457,"params":458,"resultUrl":452,"resultKind":91},"b1-p","Prompt: dolly-zoom (dolly zoom (vertigo))","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single move): as the person steps back, the camera rushes backward with them with a strong dolly-zoom (vertigo) effect, the background stretching behind them, starting on the very first frame.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":460,"type":56,"text":461,"level":82,"eyebrow":462},"b6-h","Plongée de cima (top-down crane down)","19",{"id":464,"type":32,"markdown":465},"b6-t","A câmera começa no alto, olhando reto para baixo, e vai descendo até a altura dos olhos. Mostra o lugar inteiro antes de chegar em você.",{"id":467,"type":89,"before":468,"after":471,"beforeLabel":97,"afterLabel":98},"b6-c",{"kind":91,"url":469,"posterUrl":470},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F0776b747-fe8b-48f5-a85c-055cf5814f6b.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fa562cdad-984e-479a-a9ab-520327620eab.jpg",{"kind":91,"url":472,"posterUrl":473},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Ff3b8231e-edd5-4d3d-909f-1991bb93cdb2.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F56db9104-b7bb-4b7d-be06-b866fcbcc9d8.jpg",{"id":475,"type":14,"title":476,"studio":91,"modelKey":102,"modelName":103,"prompt":477,"params":478,"resultUrl":472,"resultKind":91},"b6-p","Prompt: plongée de cima (top-down crane down)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA, calm and elegant, one continuous take, smooth motion-control feel, level horizon, no shake: a high wide crane shot looking straight down from above, slowly descending toward eye level, revealing the place around the person.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":480,"type":56,"text":481,"level":82,"eyebrow":482},"m18-h","Close nas mãos (close-up on the hands)","20",{"id":484,"type":32,"markdown":485},"m18-t","Um plano fechado nas mãos enquanto você gesticula. Serve para mostrar que o gesto continua igual ao original.",{"id":487,"type":89,"before":488,"after":491,"beforeLabel":97,"afterLabel":98},"m18-c",{"kind":91,"url":489,"posterUrl":490},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F7d6500f3-56b3-4eec-a623-cdd9dd4fe3b7.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F78a03579-db20-471a-b31b-16635f49461e.jpg",{"kind":91,"url":492,"posterUrl":493},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Ffbceb9de-ca50-4b2e-8ac1-b48f47066956.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F851a8f5c-ebb1-481b-ac4a-495ee8b2cdec.jpg",{"id":495,"type":14,"title":496,"studio":91,"modelKey":102,"modelName":103,"prompt":497,"params":498,"resultUrl":492,"resultKind":91},"m18-p","Prompt: close nas mãos (close-up on the hands)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single framing, from the very first frame and held to the end): CLOSE-UP ON THE HANDS: tight shot of the hands gesturing at chest height, the chin at the top edge of the frame, shallow depth of field.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":500,"type":56,"text":501,"level":82,"eyebrow":502},"m19-h","Close na boca (extreme close-up on the mouth)","21",{"id":504,"type":32,"markdown":505},"m19-t","Um close bem fechado no rosto, do nariz ao queixo. Serve para mostrar a boca no tempo da fala.",{"id":507,"type":89,"before":508,"after":511,"beforeLabel":97,"afterLabel":98},"m19-c",{"kind":91,"url":509,"posterUrl":510},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fc707ff77-bf5f-4b25-b54c-b9e6ef6e2c30.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fdfa864df-259a-4419-99f2-5b20c5939026.jpg",{"kind":91,"url":512,"posterUrl":513},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F3797c1f3-d9dc-45aa-bf73-b12e57cd7330.mp4","https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fe6f0ee5a-af03-491e-b42a-327f0a8ee5bd.jpg",{"id":515,"type":14,"title":516,"studio":91,"modelKey":102,"modelName":103,"prompt":517,"params":518,"resultUrl":512,"resultKind":91},"m19-p","Prompt: close na boca (extreme close-up on the mouth)","The same take, re-photographed by a different camera. The ONLY thing that changes is the camera.\n\nTHE PERFORMANCE IS 1:1: second N of the output is second N of the source. Nothing is retimed, sped up, slowed down, frozen or re-acted. Every step, hand gesture and head turn happens at the same moment, with the same hand, as in the source. The mouth stays in perfect sync with the original audio.\n\nCAMERA (one single framing, from the very first frame and held to the end): EXTREME CLOSE-UP ON THE MOUTH: from the nose to the chin, the mouth filling the frame, talking in perfect sync, shallow depth of field.\n\nNone of this may appear: the camera move starting late; a gesture missing, early or late; mouth out of sync; a different face; a cut; new people; camera equipment; text; logos; music.",{"aspectRatio":106,"resolution":107},{"id":520,"type":56,"text":521,"level":58},"hd-h","Do rascunho em 480p ao vídeo em 1080p",{"id":523,"type":32,"markdown":524},"hd-t","Depois de aprovar todos os movimentos no rascunho, eu refiz cada trecho em 1080p no Seedance 2.5 Edit usando o próprio rascunho como vídeo de origem. A câmera, o tempo e os gestos saem iguais ao rascunho, só mais nítidos. Este é o prompt que eu usei para isso:",{"id":526,"type":14,"title":527,"studio":91,"modelKey":102,"modelName":103,"prompt":528,"params":529},"hd-p","Prompt para refazer o rascunho em 1080p","The source video is the approved low-resolution draft of this exact shot. Re-render it in high resolution, 1:1: the same camera move, the same framing at every second, the same timing, the same gestures, the same mouth movements, the same face, the same clothes and the same location. Nothing is added, removed, moved, retimed or re-staged. Only the image quality improves: sharp detail, natural skin texture, clean natural light.\n\nNone of this may appear: any change of camera move, framing, timing, gesture, mouth, face, clothes or background compared with the source video; a cut; new people; text; logos; music.",{"aspectRatio":106,"resolution":530},"1080p",{"id":532,"type":56,"text":533,"level":58},"usado-h","O que foi usado",{"id":535,"type":32,"markdown":536},"usado-t","Um take de 60 segundos gravado no quintal com o celular parado, o Seedance 2.5 Edit da KAIROGEN em 480p para o rascunho e em 1080p para o final, e o Claude conectado na KAIROGEN pelo MCP para cortar os trechos e mandar as gerações. Para trocar a luz do mesmo take, veja os [tipos de iluminação cinematográfica para o prompt](\u002Fblog\u002Ftipos-de-iluminacao-cinematografica-prompt). Para mudar o ângulo e o lugar, veja o guia [vídeo parado vira dinâmico com vários ângulos](\u002Fblog\u002Fvideo-parado-vira-dinamico-varios-angulos).",{"id":538,"type":539,"name":540,"description":541,"command":542,"url":543},"sk1","skill","MCP da KAIROGEN no Claude","Conecta o Claude à KAIROGEN para gerar imagem e vídeo direto da conversa.","https:\u002F\u002Fmcp.kairogen.ai\u002Fmcp","https:\u002F\u002Fapp.kairogen.ai\u002Fmcp",{"id":545,"type":545,"title":546,"text":547,"buttonLabel":548,"href":549},"cta","Mude a câmera do seu vídeo","Suba um take seu no Seedance 2.5 Edit e cole o prompt do movimento que você quer.","Testar no estúdio de vídeo","\u002Fstudio\u002Fvideo?model=seedance-2-5",{"id":551,"type":56,"text":552,"level":58},"faq-h","Perguntas frequentes",{"id":554,"type":56,"text":555,"level":82},"faq1-h","Quais são os movimentos de câmera mais usados no cinema?",{"id":557,"type":32,"markdown":558},"faq1-t","Os básicos são aproximação (push-in), afastamento (pull back), traveling lateral, órbita, grua, câmera na mão e zoom. Para vídeo curto, chicote, FPV e câmera girando dão mais energia. A lista acima traz mais de 20, com o nome em inglês de cada um.",{"id":560,"type":56,"text":561,"level":82},"faq2-h","Preciso de equipamento para fazer esses movimentos de câmera com IA?",{"id":563,"type":32,"markdown":564},"faq2-t","Não. O take do exemplo foi gravado com o celular parado num tripé, num plano aberto de corpo inteiro. O Seedance 2.5 Edit refaz o mesmo take com a câmera nova e mantém o gesto, o rosto e a boca do original.",{"id":566,"type":56,"text":567,"level":82},"faq3-h","Dá para pedir vários movimentos de câmera num prompt só?",{"id":569,"type":32,"markdown":570},"faq3-t","Dá, mas sai melhor com um movimento por geração. Corte um trecho de 4 segundos para cada movimento e junte os trechos na edição, cada um no segundo em que você fala o nome dele.",{"id":572,"type":56,"text":573,"level":82},"faq4-h","Quanto custa mudar o movimento de câmera de um vídeo no Seedance 2.5 Edit?",{"id":575,"type":32,"markdown":576},"faq4-t","No meu plano, cada trecho de 4 segundos custou 43 créditos em 480p e cerca de 210 créditos em 1080p. Por isso eu aprovo tudo no rascunho em 480p e só depois passo para 1080p.",[578,579,580,581,582,583,584,585,586,587,588,589,590,591,592,593,594,595,596,597,598,599,600,601,602,603,604,605,606,607,608,609,610,611,612],{"id":55,"text":57,"level":58},{"id":64,"text":65,"level":58},{"id":77,"text":78,"level":58},{"id":80,"text":81,"level":82},{"id":109,"text":110,"level":82},{"id":129,"text":130,"level":82},{"id":149,"text":150,"level":82},{"id":173,"text":174,"level":58},{"id":176,"text":177,"level":82},{"id":196,"text":197,"level":82},{"id":216,"text":217,"level":82},{"id":236,"text":237,"level":82},{"id":251,"text":252,"level":82},{"id":271,"text":272,"level":58},{"id":274,"text":275,"level":82},{"id":294,"text":295,"level":82},{"id":314,"text":315,"level":82},{"id":334,"text":335,"level":82},{"id":354,"text":355,"level":58},{"id":357,"text":358,"level":82},{"id":377,"text":378,"level":82},{"id":397,"text":398,"level":82},{"id":417,"text":418,"level":82},{"id":437,"text":438,"level":58},{"id":440,"text":441,"level":82},{"id":460,"text":461,"level":82},{"id":480,"text":481,"level":82},{"id":500,"text":501,"level":82},{"id":520,"text":521,"level":58},{"id":532,"text":533,"level":58},{"id":551,"text":552,"level":58},{"id":554,"text":555,"level":82},{"id":560,"text":561,"level":82},{"id":566,"text":567,"level":82},{"id":572,"text":573,"level":82},[102],[540],"2026-11-11T00:06:33.201Z",{"title":617,"description":618,"noindex":28},"Movimentos de câmera para IA: 20 comandos com os prompts","Movimentos de câmera para IA: os nomes em português e inglês, para que serve cada um e o prompt que refaz o seu vídeo no Seedance 2.5 Edit.",[620,642,657],{"id":621,"slug":622,"title":623,"excerpt":624,"category":9,"tags":625,"cover":636,"author":639,"readingMinutes":640,"publishedAt":641,"featured":28},"6af354a6cea2ae551585cfec","criar-videos-sem-aparecer-com-ia","Criar vídeos sem aparecer com IA: os perfis faceless com milhões de views e como fazer o seu","Perfis sem rosto estão somando milhões de visualizações no Instagram com histórias em episódios, e vários começaram há poucas semanas. Veja quem são, por que o formato funciona e como fazer o seu a partir de uma frase, com a Fábrica Viral da KAIROGEN e o Claude.",[626,627,628,629,630,631,632,633,634,15,635],"criar vídeos sem aparecer","perfil faceless","canal faceless","canal dark com ia","vídeo faceless","fábrica viral","vídeos virais","instagram","reels","mcp",{"kind":19,"url":637,"alt":638},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fcbd49a66-2719-409e-a0fe-048f8ad042c7.jpg","Criar vídeos sem aparecer com IA: seis historinhas faceless em cartões verticais (série 2D, novelinha de objetos, E se a gravidade dobrasse, anime) e a faixa 1 FRASE VIRA HISTORINHA",{"name":23,"handle":24,"avatarUrl":25},12,"2026-11-10T17:53:20.524Z",{"id":643,"slug":644,"title":645,"excerpt":646,"category":9,"tags":647,"cover":651,"author":654,"readingMinutes":655,"publishedAt":656,"featured":28},"6aef9c380a2a926de96ba683","criar-ugc-com-ia-persona-de-marca","Criar UGC com IA: um take vira as personas da sua marca no Seedance 2.5 Edit (com os prompts)","Gravei um vídeo só, na sala, e virei um homem de 70 anos, um Gen Z, uma modelo de vestido verde e o garoto-propaganda de terno, até falando espanhol. Aqui estão os prompts de cada persona no Seedance 2.5 Edit da KAIROGEN.",[648,649,13,650,17],"ugc com ia","persona de marca","criativo ugc",{"kind":19,"url":652,"alt":653},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002Fe627e617-4ffc-420f-98e8-43991c0a60e6.jpg","Criar UGC com IA: o mesmo take na sala, original, envelhecido com barba branca, com visual Gen Z e como uma modelo de vestido verde num terraço. Texto: PERSONAS DA MARCA",{"name":23,"handle":24,"avatarUrl":25},5,"2026-11-07T22:58:23.326Z",{"id":658,"slug":659,"title":660,"excerpt":661,"category":9,"tags":662,"cover":664,"author":667,"readingMinutes":668,"publishedAt":669,"featured":28},"6aef9c2b0a2a926de96ba56c","tipos-de-iluminacao-cinematografica-prompt","Tipos de iluminação cinematográfica: 34 luzes para colar no prompt do Seedance 2.5 Edit (com os prompts)","Gravei um take só, no sofá, com a luz da sala, e pedi a mesma cena com 34 luzes de cinema no Seedance 2.5 Edit da KAIROGEN. Aqui está a lista com os 34 nomes prontos para colar no prompt e os prompts que eu usei.",[663,13,14,16,17],"iluminação cinematográfica",{"kind":19,"url":665,"alt":666},"https:\u002F\u002Fcdn.kairogen.ai\u002Fblog\u002F2026\u002F11\u002F4471ccca-51dc-4645-baa8-075c15b5486a.jpg","Tipos de iluminação cinematográfica: o mesmo take no sofá com luz neon, vermelho total, luz volumétrica e hora dourada, e o original pequeno no canto. Texto: 34 LUZES DE CINEMA",{"name":23,"handle":24,"avatarUrl":25},6,"2026-11-07T22:58:22.767Z",0]