图像识别

POST

/kling/v1/videos/image-recognize

最后修改时间：2026-11-05 08:06:56

识别一张图片的内容并生成自然语言描述。输入图片可以是可访问的 URL 或 Base64 编码字符串。
支持 JPG/JPEG/PNG；文件大小不超过 10MB；图片宽高均应不小于 300px；宽高比应在 1:2.5 至 2.5:1 范围内。
官方完整文档：https://www.klingai.com/document-api/api/video/image-recognition

请求参数

Header 参数

Body 参数application/json必填

示例

{
    "image": "https://h2.inkwai.com/bs2/upload-ylab-stunt/example.jpg"
}

请求示例代码

Shell

JavaScript

Java

Swift

PHP

Python

HTTP

Objective-C

Ruby

OCaml

Dart

curl --location '/kling/v1/videos/image-recognize' \
--header 'Authorization: Bearer {{YOUR_API_KEY}}' \
--header 'Content-Type: application/json' \
--data '{
    "image": "https://h2.inkwai.com/bs2/upload-ylab-stunt/example.jpg"
}'

返回响应

🟢200

application/json

识别成功

Bodyapplication/json

示例

{
    "code": 0,
    "message": "string",
    "request_id": "string",
    "data": {
        "task_result": {
            "images": [
                {
                    "type": "object_seg",
                    "is_contain": true,
                    "url": "string"
                },
                {
                    "type": "head_seg",
                    "is_contain": true,
                    "url": "string"
                },
                {
                    "type": "face_seg",
                    "is_contain": true,
                    "url": "string"
                },
                {
                    "type": "cloth_seg",
                    "is_contain": true,
                    "url": "string"
                }
            ]
        },
        "final_unit_deduction": "string",
        "final_balance_deduction": {
            "quota": "string",
            "list_price": "string"
        }
    }
}

🟠400

🟠401

修改于 2026-11-05 08:06:56

创建对口型任务

Blend提交参数