Sheru Bhai AI

Sheru Bhai AI 视频生成器

一张自拍,和 Sheru 一起当主角。

了解这个工具

上传自己的照片,走进 Sheru Bhai AI 视频:和白狮并肩登场、撒钱庆祝、握手成交,或从本人变身白狮 Don。每个玩法都配有精细的 Sheru Bhai AI video prompt,直接使用 Wan 3 生成。

1. 选择你和 Sheru 的玩法

2. 上传你的自拍

四种玩法都必须上传一张照片。你的面容、发型和服装只替换原片里的示例真人,身旁的白狮 Sheru 保持原样。

你的自拍必填

使用一位成年人的清晰照片,面部无遮挡,并展示你想穿的衣服,不限成人性别。JPG、PNG 或 WebP,≤10 MiB,至少 300 × 300 像素。AI 可能改变相似度或手部细节。

输出设置
7 秒 · 9:16 竖屏输出 · 保留参考原片声音。
输出时长

7 秒

完整场景共 280 积分,失败或超时按原实扣金额退款。

看看你的照片会出现在这里

来自 aisheru.com 的 AI 参考原片。生成时用你的照片替换所选场景中的示例真人;此预览非 ImageLayered 或 Wan 3 实测结果。

将被替换的示例人物

将被替换的示例人物

参考站 AI 示例肖像,上传自己的照片即可接替这一角色。

  1. 01选择四种照片玩法之一。
  2. 02上传自己的肖像,仅替换当前场景指定的真人角色。
  3. 03分享前检查你的身份、与 Sheru 的互动和声音。
7 秒
480p
9:16
原片声音

一张照片,四种和 Sheru 一起玩的方式

每组展示参考站的 AI 示例肖像与对应视频,素材来自 aisheru.com。参考站标注其视频由 MiniMax H3 通过 YIR 制作,非 ImageLayered 或 Wan 3 实测输出。本站四种玩法均需上传自己的照片。

参考输入照片

参考输入照片

参考场景 · 生成时换成你

并肩登场

从酒店走出,和白狮并肩走向镜头。

必填 1 张照片 · 7 秒

来源:aisheru.com/video · 导入 AI 示例
参考输入照片

参考输入照片

参考场景 · 生成时换成你

撒钱庆祝

你抛起纸钞,Sheru 在身旁大笑。

必填 1 张照片 · 7 秒

来源:aisheru.com/video · 导入 AI 示例
参考输入照片

参考输入照片

参考场景 · 生成时换成你

握手成交

坐到对面,和白狮 Don 握手庆祝。

必填 1 张照片 · 7 秒

来源:aisheru.com/video · 导入 AI 示例
参考输入照片

参考输入照片

参考场景 · 生成时换成你

变身白狮

从你的自拍,过渡为白西装白狮。

必填 1 张照片 · 7 秒

来源:aisheru.com/video · 导入 AI 示例

每个剧情都有精细的 Sheru Bhai AI video prompt

提示词将照片绑定到准确的真人角色:并肩登场替换左侧真人,撒钱庆祝替换右侧真人,握手成交替换左侧真人;白狮 Sheru 的脸和白西装保持原样。变身版单独约束从本人到狮子的过程。下方可免费查看、复制完整 prompt,生成保留原片声音。

并肩登场

Image 1 identifies the ONE human participant to insert into Video 1. Match the reference person's facial structure, eye shape and spacing, brows, nose, mouth, skin tone, apparent adult age, hairline, hairstyle, hair length, facial hair and visible personal accessories. Keep this one identity stable through profiles, motion, smiles, partial occlusions and changing camera distance. Replace the assigned sample human's entire visible appearance, hair, body proportions and outfit with the reference person's appearance and clothes, adapting the outfit naturally to the original poses. If trousers or shoes are absent from Image 1, extend the visible outfit coherently with plain neutral lower garments. Do not import the photo background. Preserve the source performance and gaze using the reference person's own anatomy. No pasted face, face blending, residual sample-person features, identity flicker or duplicate uploaded people. The reference person may be any adult gender; do not inherit the sample actor's gender or body shape. BOSS ENTRY, ONE CONTINUOUS ENTRANCE: Replace ONLY the dark-haired adult woman on the LEFT of the white lion, wearing a black short-sleeve top and beige trousers in the source. Image 1 becomes this human participant. At 0-2s the pair step out through the hotel's gold-framed glass doors; at 2-4s they walk toward the camera side by side; at 4-6.58s the camera sees them closer, smiling, still in their original left/right order. Keep the uploaded person's own outfit throughout, preserve their unhurried stride and natural arm swing, and keep their head and feet connected to the same body. Sheru stays to the human's RIGHT. Keep the door frame, chandelier lobby, entrance floor and camera approach. Exactly one reference-derived HUMAN and one original WHITE LION; never make a second Sheru or exchange their faces. PROTECTED SHERU: the large adult anthropomorphic white lion, full ivory mane, gold chain and wristwatch, tailored white three-piece suit. Preserve his exact source lion face, white fur, mane, muzzle, costume, accessories, body scale, gestures and expressions. Do NOT apply Image 1 to Sheru, turn him into the uploaded human, copy his lion head onto the human participant or swap their roles. Screen position is only a starting cue; follow the original human's identity throughout. Keep background bystanders unchanged. Edit the supplied Video 1; do not generate an unrelated scene. Preserve its original shot order, actual cuts, camera motion, framing, gestures, timing, light, shadows, furnishings, background people and prop trajectories. Generate seven seconds: follow the complete approximately 6.58-second source, then briefly hold the natural ending for the remaining fraction; do not add a new cut, plot beat or extra action. Keep both featured characters in the portrait framing, with their original positions and relative scale. Do not crop either face at the frame edges. Preserve the original audio track and its synchronization; do not invent new dialogue, a new speaker, new music or subtitles. Keep hands anatomically coherent, grip contact plausible, feet grounded, clothing temporally stable, reflections consistent, and objects in front of the correct person. No extra limbs, merged hands, transparent bodies, extra uploaded identities, added text, logos or watermarks.

撒钱庆祝

Image 1 identifies the ONE human participant to insert into Video 1. Match the reference person's facial structure, eye shape and spacing, brows, nose, mouth, skin tone, apparent adult age, hairline, hairstyle, hair length, facial hair and visible personal accessories. Keep this one identity stable through profiles, motion, smiles, partial occlusions and changing camera distance. Replace the assigned sample human's entire visible appearance, hair, body proportions and outfit with the reference person's appearance and clothes, adapting the outfit naturally to the original poses. If trousers or shoes are absent from Image 1, extend the visible outfit coherently with plain neutral lower garments. Do not import the photo background. Preserve the source performance and gaze using the reference person's own anatomy. No pasted face, face blending, residual sample-person features, identity flicker or duplicate uploaded people. The reference person may be any adult gender; do not inherit the sample actor's gender or body shape. MONEY RAIN, ONE CONTINUOUS CELEBRATION: Replace ONLY the dark-haired adult woman in the black top standing on the RIGHT of Sheru, behind the foreground table. Image 1 is this human participant. At 0-1.5s the human holds a stack of fictional notes in both hands near waist level; around 1.5-2.5s the human lifts both arms and tosses the notes upward; around 2.5-4.5s the notes fall in front of and behind both characters while the human looks up smiling and the original lion laughs; around 4.5-6.58s the human lowers their arms as notes settle. Preserve the upload's identity even when notes briefly cover the face or hair, and re-reveal the SAME person afterward. Sheru stays on the LEFT, with his original lion face and white suit. Preserve the golden lantern street, bystanders, table, dishes and note movement; do not replace them with an office or invent readable currency designs. Keep the original throw trajectory and finger positions, without fused hands or banknotes glued to skin. PROTECTED SHERU: the large adult anthropomorphic white lion, full ivory mane, gold chain and wristwatch, tailored white three-piece suit. Preserve his exact source lion face, white fur, mane, muzzle, costume, accessories, body scale, gestures and expressions. Do NOT apply Image 1 to Sheru, turn him into the uploaded human, copy his lion head onto the human participant or swap their roles. Screen position is only a starting cue; follow the original human's identity throughout. Keep background bystanders unchanged. Edit the supplied Video 1; do not generate an unrelated scene. Preserve its original shot order, actual cuts, camera motion, framing, gestures, timing, light, shadows, furnishings, background people and prop trajectories. Generate seven seconds: follow the complete approximately 6.58-second source, then briefly hold the natural ending for the remaining fraction; do not add a new cut, plot beat or extra action. Keep both featured characters in the portrait framing, with their original positions and relative scale. Do not crop either face at the frame edges. Preserve the original audio track and its synchronization; do not invent new dialogue, a new speaker, new music or subtitles. Keep hands anatomically coherent, grip contact plausible, feet grounded, clothing temporally stable, reflections consistent, and objects in front of the correct person. No extra limbs, merged hands, transparent bodies, extra uploaded identities, added text, logos or watermarks.

握手成交

Image 1 identifies the ONE human participant to insert into Video 1. Match the reference person's facial structure, eye shape and spacing, brows, nose, mouth, skin tone, apparent adult age, hairline, hairstyle, hair length, facial hair and visible personal accessories. Keep this one identity stable through profiles, motion, smiles, partial occlusions and changing camera distance. Replace the assigned sample human's entire visible appearance, hair, body proportions and outfit with the reference person's appearance and clothes, adapting the outfit naturally to the original poses. If trousers or shoes are absent from Image 1, extend the visible outfit coherently with plain neutral lower garments. Do not import the photo background. Preserve the source performance and gaze using the reference person's own anatomy. No pasted face, face blending, residual sample-person features, identity flicker or duplicate uploaded people. The reference person may be any adult gender; do not inherit the sample actor's gender or body shape. DEAL DONE, ONE SEATED TWO-SHOT: Replace ONLY the adult bearded dark-haired man in the burgundy long-sleeve shirt and light trousers seated on the LEFT, seen mainly in profile. Image 1 becomes this HUMAN; maintain its own facial geometry in profile rather than retaining the sample man's beard. Keep the uploaded person's own clothes, seated posture, gaze toward the lion and original arm position. At 0-1s the two lean forward into the right-hand handshake; at 1-4s they clasp hands naturally across the gap between the chairs; around 4-6.58s the original lion on the RIGHT lifts his head and opens his mouth in a celebratory roar while the human remains seated. Protect the lion's entire face, mane, suit and roar. Match only the human side of the handshake to Image 1, keeping exactly two separate hands with a plausible grip; do not replace the lion paw or grow another arm. Preserve both office chairs, desk behind them, chandelier, framed pictures, floor reflections and the stationary camera. PROTECTED SHERU: the large adult anthropomorphic white lion, full ivory mane, gold chain and wristwatch, tailored white three-piece suit. Preserve his exact source lion face, white fur, mane, muzzle, costume, accessories, body scale, gestures and expressions. Do NOT apply Image 1 to Sheru, turn him into the uploaded human, copy his lion head onto the human participant or swap their roles. Screen position is only a starting cue; follow the original human's identity throughout. Keep background bystanders unchanged. Edit the supplied Video 1; do not generate an unrelated scene. Preserve its original shot order, actual cuts, camera motion, framing, gestures, timing, light, shadows, furnishings, background people and prop trajectories. Generate seven seconds: follow the complete approximately 6.58-second source, then briefly hold the natural ending for the remaining fraction; do not add a new cut, plot beat or extra action. Keep both featured characters in the portrait framing, with their original positions and relative scale. Do not crop either face at the frame edges. Preserve the original audio track and its synchronization; do not invent new dialogue, a new speaker, new music or subtitles. Keep hands anatomically coherent, grip contact plausible, feet grounded, clothing temporally stable, reflections consistent, and objects in front of the correct person. No extra limbs, merged hands, transparent bodies, extra uploaded identities, added text, logos or watermarks.

变身白狮

Image 1 is the single uploaded adult identity for the human phase ONLY. Follow the transformation scope below. SHERU TRANSFORM, ONE PERSON BECOMES ONE LION: Image 1 replaces ONLY the initial human selfie subject and that human's visible appearance before and during the transformation. At approximately 0-0.4s show the uploaded person, recognizable, at the source market-street selfie angle with the original outstretched arm, using their own hair and clothes. Around 0.4-1.5s follow the source's magical golden smoke transformation: the SAME head progressively develops a white-lion muzzle, ivory fur and mane while the person's body, selfie pose and reference clothes remain connected. This is a deliberate gradual identity transformation, not a face-swap error; do not show a different sample human, a detached lion mask or two simultaneous people. Around 1.5-2s preserve the actual source cut to the palace-room full-body WHITE LION in the white three-piece suit and gold jewelry. From approximately 2-6.58s preserve that FINAL LION's original face, ivory mane, suit, body gestures and mouth movement. The final lion is the fictional Sheru, not a human wearing the uploaded person's face; do not reapply Image 1 after the transformation is complete. Preserve the original street-to-palace scene transition and the original audio; no extra transition, second companion or duplicated body. At all times exactly ONE featured character, first human, then transforming, then lion. Edit the supplied Video 1; do not generate an unrelated scene. Preserve its original shot order, actual cuts, camera motion, framing, gestures, timing, light, shadows, furnishings, background people and prop trajectories. Generate seven seconds: follow the complete approximately 6.58-second source, then briefly hold the natural ending for the remaining fraction; do not add a new cut, plot beat or extra action. Keep both featured characters in the portrait framing, with their original positions and relative scale. Do not crop either face at the frame edges. Preserve the original audio track and its synchronization; do not invent new dialogue, a new speaker, new music or subtitles. Keep hands anatomically coherent, grip contact plausible, feet grounded, clothing temporally stable, reflections consistent, and objects in front of the correct person. No extra limbs, merged hands, transparent bodies, extra uploaded identities, added text, logos or watermarks.

Who started the Sheru Bhai trend?

Sheru Bhai 是多个 AI 短片中出现的虚构白狮 Don。截至 2026 年 11 月 16 日本页核查,来源尚不足以独立确认唯一首创者。参考站讨论了 @digitalsheru、Digital Sheru 等不同归属说法,不能仅凭这些自述确定谁上传了第一条视频。

如何用自己的照片制作 Sheru Bhai AI 视频

选好和 Sheru 的互动,上传自拍,进入画面。

1. 选择你和 Sheru 的玩法

和白狮并肩登场、撒钱庆祝、握手成交,或直接变身白狮。每个模板都使用你自己的照片。

2. 上传一张自拍

上传一位成年人的清晰肖像,选择 480p、720p 或 1080p,查看积分报价,再用 Wan 3 生成。不用自己编写提示词。

3. 玩游戏、检查并下载

等待时玩 2048。检查你的身份、手部、互动和原片声音后下载 MP4,完成结果保留 30 天。

Sheru Bhai AI 常见问题

了解照片、人物位置、提示词、积分、声音和任务恢复。

Sheru Bhai AI 可以怎么玩?

上传一张自己的照片,和虚构的白狮 Don 互动。「并肩登场」和他一起走出酒店;「撒钱庆祝」由你抛起纸钞;「握手成交」让你坐在对面与他握手;「变身白狮」则从你的自拍开始,变为白狮。

Sheru Bhai AI video prompt 怎样使用我的照片?

Image 1 指定用户真人身份,而不是白狮。每套提示词都写明原片人物、动作及身份约束,请求在真人角色出现时保持你的面容、发型和服装,同时保护 Sheru。变身版仅在变化前及过程中使用你的身份。所有完整提示词均可免费查看和复制。

Who started the Sheru Bhai trend,谁发起了这个趋势?

截至本页核查,尚无独立证据确认唯一首创者。参考站列出多个归属说法,包括 @digitalsheru 和 Digital Sheru 的品牌故事。这些属于来源自述或待核实说法,不能单独证明首发归属;可查看起源介绍中的来源链接。

每个模板都要上传自己的照片吗?

是的,当前四个模板都必须上传一张照片。请选择面部清晰无遮挡的成年人肖像,真人互动版可展示你想穿的衣服。切换模板会保留已选照片,可用同一张照片尝试不同玩法。

Sheru Bhai AI video generator 需要多少积分?

完整 7 秒视频在 480p 下为 280 积分,720p 为 560,1080p 为 1120,照片不额外收费。约 6.58 秒的输入按 7 秒计费,再加 7 秒输出。失败或超时按原实扣金额退款。

会保留我的外观和原片声音吗?

三个同框玩法请求保留你的面容、发型、配饰与服装,白狮不变。变身版会有意将你变为狮子,最终采用 Sheru 的白西装。保留参考原片音轨,不提供自定义台词。AI 可能改变相似度、手部或动作,请检查完整结果。

等待时可以做什么?

上传及免费排队阶段请保持页面打开。任务受理后在后台继续执行,可玩本地 2048 小游戏,也可在 24 小时内恢复最新任务;已完成视频在「最近生成」保留 30 天。

和 Sheru Bhai 一起高调登场

一张自拍,让你进入画面。选好互动,和白狮 Don 一起玩。

制作我和 Sheru 的视频