video-from-image

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Video from Image

图片转视频

Animate a single image into a dynamic video using Google DeepMind's Veo 3.1 model via fal.ai.

See

references/fal-api.md

for setup, Python patterns, and error handling.

使用fal.ai调用Google DeepMind的Veo 3.1模型，将单张图片制作成动态视频。

如需了解设置方法、Python使用模式以及错误处理，请查看

references/fal-api.md

。

API Endpoint

API接口

fal-ai/veo3.1/image-to-video

fal-ai/veo3.1/image-to-video

Parameters

参数

Required

必填参数

```
prompt
```
(string): Text description of the video motion and action to generate
```
image_url
```
(string): URL of the input image to animate (use
```
fal_client.upload_file()
```
for local files)

```
prompt
```
（字符串）：用于描述要生成的视频的运动和动作的文本说明
```
image_url
```
（字符串）：待动画处理的输入图片的URL（本地文件可使用
```
fal_client.upload_file()
```
上传）

Optional

可选参数

Parameter	Type	Default	Options
`duration`	string	"8s"	"4s", "6s", "8s"
`aspect_ratio`	string	"auto"	"auto", "9:16", "16:9"
`resolution`	string	"720p"	"720p", "1080p"
`generate_audio`	boolean	true	Disable to save ~50% credits

参数	类型	默认值	可选值
`duration`	字符串	"8s"	"4s", "6s", "8s"
`aspect_ratio`	字符串	"auto"	"auto", "9:16", "16:9"
`resolution`	字符串	"720p"	"720p", "1080p"
`generate_audio`	布尔值	true	关闭该选项可节省约50%的积分

CLI Script

CLI脚本

bash

python3 scripts/fal_generate.py \
    --endpoint video-from-image \
    --prompt "Camera slowly pans across the landscape" \
    --image /path/to/photo.jpg \
    --duration 8s \
    --video-resolution 1080p \
    --output video.mp4

bash

python3 scripts/fal_generate.py \
    --endpoint video-from-image \
    --prompt "Camera slowly pans across the landscape" \
    --image /path/to/photo.jpg \
    --duration 8s \
    --video-resolution 1080p \
    --output video.mp4

Tips

小贴士

Use descriptive motion language in prompts: pan, zoom, sway, flow, drift
Describe camera movements for cinematic effects: "slow dolly forward", "aerial pull back"
Higher resolution (1080p) gives better quality but takes longer and costs more
Disable audio generation (
```
generate_audio: False
```
) to save credits when sound is not needed
Ensure input image is at least 720p for best results
Shorter duration (4s) generates faster and is good for quick previews

在prompt中使用描述性的动作词汇：平移（pan）、缩放（zoom）、摇摆（sway）、流动（flow）、漂移（drift）
描述相机运动以实现电影级效果：“缓慢向前推镜头（slow dolly forward）”、“空中拉远镜头（aerial pull back）”
更高分辨率（1080p）画质更好，但生成时间更长、成本更高
当不需要声音时，关闭音频生成（
```
generate_audio: False
```
）以节省积分
确保输入图片分辨率至少为720p以获得最佳效果
更短的时长（4s）生成速度更快，适合快速预览