Available in VPC
Get the progress and results of the requested conversion job. If the conversion status is TRANSFORM_COMPLETED, the response will include the conversion results for each preset (result.output). Since conversions are processed asynchronously, use the conversion job ID returned by the Request scene summary utilization to check the status until it reaches the completed state.
The structure of the conversion result (result.output) varies depending on the requested preset. Distinguish the result structure based on the result.preset value.
Request
This section describes the request format. The method and URI are as follows:
| Method | URI |
|---|---|
| GET | /api/v1/workspaces/{workspace_name}/projects/{project_id}/assets/{asset_id}/scene-summary/transformations/{transformation_id} |
Request headers
For information about the headers common to all Media Intelligence APIs, see Media Intelligence request headers.
Request path parameters
The following describes the parameters.
| Field | Type | Required | Description |
|---|---|---|---|
workspace_name |
String | Required | Workspace name |
project_id |
String | Required | Project ID
|
asset_id |
String | Required | Media asset ID
|
transformation_id |
String | Required | Conversion job ID |
Request example
The request example is as follows:
curl --location --request GET 'https://mi.apigw.ntruss.com/api/v1/workspaces/my-workspace/projects/1234/assets/5678/scene-summary/transformations/2001' \
--header 'x-ncp-apigw-timestamp: {Timestamp}' \
--header 'x-ncp-iam-access-key: {Access Key}' \
--header 'x-ncp-apigw-signature-v2: {API Gateway Signature}' \
--header 'Content-Type: application/json'
Response
This section describes the response format.
Response body
The response body includes the following data:
| Field | Type | Required | Description |
|---|---|---|---|
code |
String | - | API processing result code |
message |
String | - | API processing result message |
result |
Object | - | Conversion status and results information |
result.status |
String | - | Conversion status
|
result.transformStatusMessage |
String | - | Detailed error message when conversion fails
|
result.preset |
String | - | Used conversion preset
|
result.locale |
String | - | Conversion result language
|
result.indexId |
Integer | - | Index ID of the scene summary used for the conversion
|
result.presetVersion |
String | - | Preset conversion policy version
|
result.engineVersion |
String | - | Coversion engine version
|
result.outputFields |
Array<String> | - | Applied output field
|
result.keywordCount |
Integer | - | Number of keywords per applied scene
|
result.paragraphCount |
Integer | - | Number of summary paragraphs per applied scene
|
result.usage |
Object | - | Usage information
|
result.output |
Object | - | Conversion results by preset
|
usage
The following describes usage.
| Field | Type | Required | Description |
|---|---|---|---|
outputTokens |
Integer | - | Number of output tokens used for billing |
output (SCENE_DIGEST)
The following describes the output when preset is set to SCENE_DIGEST.
| Field | Type | Required | Description |
|---|---|---|---|
scenes |
Array | - | List of summary results by scene |
scenes[].scene |
Integer | - | Scene ID |
scenes[].startMs |
Integer | - | Scene start time (millisecond) |
scenes[].endMs |
Integer | - | Scene end time (millisecond) |
scenes[].title |
String | - | Representative scene title
|
scenes[].keywords |
Array<String> | - | List of key scene keywords
|
scenes[].paragraphs |
Array<String> | - | List of scene summary paragraphs
|
output (SCENE_METADATA)
The following describes the output when preset is set to SCENE_METADATA.
| Field | Type | Required | Description |
|---|---|---|---|
scenes |
Array | - | List of metadata by scene |
scenes[].scene |
Integer | - | Scene ID |
scenes[].startMs |
Integer | - | Scene start time (millisecond) |
scenes[].endMs |
Integer | - | Scene end time (millisecond) |
scenes[].timeRange |
Object | - | Display time and timecode informatio
|
scenes[].tags |
Array<String> | - | List of tags for searching and filtering |
scenes[].content |
Object | - | Observation-based scene metadata: content |
content
The following describes content. All subfields are optional, and information not observed in the scene is omitted. Since new fields may be added in the future, implement your code to ignore any undefined fields.
| Field | Type | Required | Description |
|---|---|---|---|
objects |
Array<String> | - | List of names of major objects in the scene
|
actions |
Array<String> | - | List of major actions observed in the scene |
environment |
Object | - | Background information
|
composition |
Object | - | Information on the shot composition
|
persons |
Array<Object> | - | List of labeled people
|
textInScene |
Array<String> | - | Original text appearing in the scene (subtitles, signs, on-screen text, etc.) |
Response status codes
For information about the HTTP status codes common to all Media Intelligence APIs, see Media Intelligence response status codes.
Response example
The following is a sample response for a conversion in progress (TRANSFORMING).
{
"code": "0",
"message": "success",
"result": {
"status": "TRANSFORMING",
"transformStatusMessage": null,
"preset": "SCENE_DIGEST",
"locale": "en-US"
}
}
The following is a sample response for a completed conversion (TRANSFORM_COMPLETED) of the SCENE_DIGEST preset.
{
"code": "0",
"message": "success",
"result": {
"status": "TRANSFORM_COMPLETED",
"transformStatusMessage": null,
"preset": "SCENE_DIGEST",
"locale": "ko-KR",
"indexId": 1001,
"presetVersion": "1.0.0",
"engineVersion": "1.0.0",
"outputFields": ["title", "keywords", "paragraphs"],
"keywordCount": 5,
"paragraphCount": 1,
"usage": {
"outputTokens": 380
},
"output": {
"scenes": [
{
"scene": 1,
"startMs": 0,
"endMs": 5000,
"title": "Host's interview opening",
"keywords": ["Gwanghwamun", "host", "interview", "square", "start"],
"paragraphs": [
"The host announces the start of a citizen interview program at Gwanghwamun Square."
]
},
{
"scene": 2,
"startMs": 5000,
"endMs": 12000,
"title": "Conducting citizen interviews",
"keywords": ["citizens", "daily life", "interview", "street", "conversation"],
"paragraphs": [
"Various citizens appear and share their daily lives and thoughts."
]
}
]
}
}
}
The following is a sample response for a completed conversion (TRANSFORM_COMPLETED) of the SCENE_METADATA preset.
{
"code": "0",
"message": "success",
"result": {
"status": "TRANSFORM_COMPLETED",
"transformStatusMessage": null,
"preset": "SCENE_METADATA",
"locale": "ko-KR",
"indexId": 1001,
"presetVersion": "1.0.0",
"engineVersion": "1.0.0",
"usage": {
"outputTokens": 640
},
"output": {
"scenes": [
{
"scene": 1,
"startMs": 1000,
"endMs": 3000,
"timeRange": {
"startTime": "00:00:01.000",
"endTime": "00:00:03.000",
"startTimecode": "00:00:01:00",
"endTimecode": "00:00:03:00",
"frameRate": 29.97
},
"tags": ["office", "laptop", "weekday", "work", "medium shot"],
"content": {
"objects": ["laptop", "desk"],
"actions": ["using laptop", "conversation"],
"environment": {
"place": "Office",
"locationType": "indoor",
"timeOfDay": "daytime"
},
"composition": {
"shotType": "medium_shot",
"cameraAngle": "eye_level"
},
"persons": [
{
"id": "person_1",
"label": "Green Kim"
}
],
"textInScene": ["Q4 REPORT"]
}
}
]
}
}
}