Black Friday

任意套餐全年立减 50%

立即领取
eroq

Tool

Image to video

Image to video AI turns a still picture into a moving clip: you bring a photo or a render, say what moves, and an engine animates it while keeping the face, the wardrobe and the framing. Every video engine on eroq takes a start frame, and the Seedance, Veo 3.1, Kling and Hailuo engines add an end frame, so a clip can travel from one exact image to another. Describe only the motion: the picture already says who and where.

视频

打开工作室并自动填好内容。生成时再注册:免费送 40 积分,无需绑卡。

视频 · 创作

10月13日 01:42:07·已完成 · 耗时 1 min 8 sec

@Mara 站在黄昏的天台上,风掀起她的大衣,脚下的城市渐次亮起灯火。变形宽银幕镜头,35mm 胶片。 推镜头

Veo 3.1 Fast · 8s · 1080p · 16:9 · 2 个版本

@Mara 站在黄昏的天台上,风掀起她的大衣,脚下的城市渐次亮起灯火。变形宽银幕镜头,35mm 胶片。 推镜头
@Mara 站在黄昏的天台上,风掀起她的大衣,脚下的城市渐次亮起灯火。变形宽银幕镜头,35mm 胶片。 推镜头
向后延伸 向前延伸
创作 Edit 新 Replace 热门 特效
添加角色

@Mara 站在黄昏的天台上,风掀起她的大衣,脚下的城市渐次亮起灯火。变形宽银幕镜头,35mm 胶片。 推镜头

Veo 3.1 Fast 语法4s6s8s
生成 816

Three ways in

  • →Set the First frame in the composer: pick a picture from your Library or Storage, or paste or drop one. On the engines that take one, a Last frame ends the clip on another picture.
  • →Add a Character with Add character: on the engines that read reference pictures their look sheet rides along, so the same face carries into the clip. Motion One does not take characters yet.
  • →Press Animate on any image in your library: it opens the video studio with that render as the source.

Prompting for motion, not for the picture

The reference already says who and where. The prompt should say what happens: "she turns toward the window and the curtain lifts in the breeze", "slow push-in as steam rises off the cup". Describing the subject again fights the reference; describing the motion directs it.

Engines and controls

Every video engine on eroq accepts a start frame. eroq Motion One, Seedance, Veo, Kling and Hailuo also accept an end frame and interpolate the motion between the two; Seedance, Veo, Runway Gen-4.5 and Wan 3.0 honor a seed for repeatable takes. Kling is the pick for faces, Hailuo 3 for physics, Motion One for uncensored, private renders. Each engine renders its own set of native aspect ratios.

Clap · AI 智能体

或者让 Clap 来做片段

在 Ask 里描述你想要的结果。Clap 会问清所需信息,挑选模型,写好提示词,并给出计划报价。你确认后,它会把所有内容生成到你的作品库。全程自动,你点头之前什么都不会运行。

  • 1Clap 一次只问一个问题,并随时更新简报、演员阵容和分镜。
  • 2运行之前,计划会列出每一项生成及其价格。
  • 3点一下“批准”,全部开始生成。失败的条目会自动退款。

Ask · Clap

三段氛围感十足的 5 秒片段:Kai 走在霓虹雨夜里,16:9。

Clap · 思考了 6 秒

同一个运镜拍三条,每条都是独立片段。计划如下。

计划 · 3 个生成任务435 积分
片段 · 场景 1 · eroq Motion One · 5s · 16:9145 积分
片段 · 场景 2 · eroq Motion One · 5s · 16:9145 积分
片段 · 场景 3 · eroq Motion One · 5s · 16:9145 积分
完成 · 全部到位

制作

已保存到你的作品库

从这里开始

选一个:它会填进顶部的输入框,直接就能生成。

隐私

私密,且不受限制

  • →eroq 自有引擎没有额外的过滤,只有一份成文政策。
  • →不保留任何内容,也不用你的作品训练模型。
  • →隐私模式不留任何文件。

常见问题

Can I animate a photo of myself?

You can animate photos of fictional characters and original creations. Real, identifiable people without consent — including deepfakes — are banned by the acceptable-use policy and enforced.

Which engines support image-to-video?

Every engine: eroq Motion One, Motion Pro, Seedance, Veo, Kling, Hailuo, Runway Gen-4.5, Grok Imagine and Wan 3.0. eroq Motion One, Seedance, Veo, Kling and Hailuo additionally take an end frame.

How do I keep the same character across clips?

Save them as a Character and @mention them in each prompt: on the engines that read reference pictures their identity sheet rides along, so the face holds. In Cinema, each scene of a film has its own cast.

Does image-to-video cost more than text-to-video?

Only on Grok Imagine, which adds 5 credits per frame picture. Everywhere else the start image is free: the price is the engine's per-second rate at your resolution, or Motion One's 29 a second.

Can I turn a photo into a video for free?

Not with the welcome credits alone: a new account gets 40 free credits, and the cheapest five-second clip is 50 credits on eroq Motion One at 480p. Credits start with a $10 pack (1,000 credits), every price shows on the Generate button before you click, and a failed render refunds itself.

What is the best AI to turn a picture into a video?

It depends on the shot. On eroq, Kling is the pick for faces, Hailuo 3 for physics, and Motion One, eroq's own engine, for uncensored and private renders. The same picture can go to several engines and the takes sit side by side, each priced before you click.

How long can a clip made from a photo be?

Up to 30 seconds, depending on the engine and your plan. For a longer video, Extend continues a clip from its last frame and Cinema chains shots into a film.

Can I animate a photo without a content filter?

Yes, on Motion One, eroq's own engine: it takes a start frame and renders what you write within the acceptable-use policy (fiction only; nothing involving minors, no real identifiable people without consent, nothing illegal). The third-party engines are moderated by their providers.

继续探索