Parameters of video AI - ApsaraVideo VOD - Alibaba Cloud Documentation Center

This topic describes parameters for video AI features, including automated review, smart thumbnail, video fingerprinting, and smart tagging.

AITemplateInfo

Parameter	Type	Required	Description
TemplateId	String	Yes	The ID of the template.
TemplateType	String	Yes	The type of the template. Valid values: AIMediaAudit: automated review template AIImage: smart thumbnail template
TemplateName	String	Yes	The name of the template.
TemplateConfig	String	Yes	The detailed configurations of the template. The value is a JSON string. For more information, see the AITemplateConfig section of this topic.
Source	String	Yes	The source of the template. Valid values: System Custom
IsDefault	String	No	Indicates whether the default template is used. Valid values: Default NotDefault
CreationTime	String	Yes	The time when the template was created. The time is displayed in UTC.
ModifyTime	String	Yes	The time when the template was modified. The time is displayed in UTC.

AITemplateConfig

Condition: TemplateType is set to AIMediaAudit

Field	Type	Required	Description
AuditItem	String[]	Yes	The item checked during automated review. Valid values: terrorism: terrorist content porn: pornographic content ad: ad violation logo:logo image live: undesirable scene audio: audio anti-spam Note If you submit a file that contains an audio track for automated review, audio review is performed only when this parameter is set to `audio`. If you submit an audio-only file for automated review, you must specify `audio` for this parameter. Do not specify `audio` for this parameter if the file that you submit for automated review does not contain an audio track.
AuditRange	String[]	Yes	The checking scope for automated review. Valid values: video image-cover text-title
AuditContent	String[]	Yes	The checking content for automated review. Valid value: screen.
AuditAutoBlock	String	Yes	Indicates whether to automatically block the video if the video fails the automated review. Valid values: no yes

Condition: TemplateType is set to AIImage

Field

Type

Required

Description

Format

String

Yes

The format of the smart thumbnail. Valid values:

SetDefaultCover

String

Yes

Indicates whether to set a thumbnail. Valid values:

true: sets an image as the video thumbnail.
false: does not set a video thumbnail.

AIMediaAuditResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for review results. Valid values: block review pass
AbnormalModules	String	No	The content that violates the regulations. Separate multiple values with commas (,). Valid values: video image-cover text-title
Label	String	Yes	The category of the content that violates the regulations in the review results. Separate multiple values with commas (,). Valid values: porn: pornographic content terrorism: terrorist or politically sensitive content ad: ad violation live: undesirable scene logo: logo image audio: audio anti-spam normal
VideoResult	VideoResult	No	The results of video review.
CoverResult Note ImageResult is returned instead of CoverResult in the response to the GetAIMediaAuditJob operation.	CoverResult[]	No	The review results of the video thumbnail.
TitleResult Note TextResult is returned instead of TitleResult in the response to the GetAIMediaAuditJob operation.	TitleResult[]	No	The review results of the video title.
AudioResult	AudioResult[]	No	The results of audio review.
ImageResult Note This field is returned only after you call the SubmitAIImageAuditJob operation.	ImageResult[]	No	The results of image review.

VideoResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for review results. Valid values: block review pass
Label	String	Yes	The category of the content that violates the regulations in the review results. Separate multiple values with commas (,). Valid values: porn: pornographic content terrorism: terrorist or politically sensitive content ad: ad violation live: undesirable scene logo: logo image normal
TerrorismResult	TerrorismResult[]	No	The results of terrorist content review.
PornResult	PornResult[]	No	The results of pornographic content review.
AdResult	AdResult[]	No	The results of ad review.
LiveResult	LiveResult[]	No	The results of undesired content review.
LogoResult	LogoResult[]	No	The results of logo review.

TerrorismResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for terrorist content review results. Valid values: block review pass
Label	String	Yes	The category of terrorist content review results. Valid values: normal bloody explosion outfit logo weapon politics violence crowd parade carcrash flag location others
MaxScore	String	Yes	The highest score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
AverageScore	String	Yes	The average score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
CounterList	TerrorismCounterList[]	Yes	The categories of terrorist content review results and the number of video snapshots in each category.
TopList	TerrorismTopList[]	Yes	The information about the top 10 video snapshots that have the highest scores in the category.

TerrorismCounterList

Field

Type

Required

Description

Label

String

Yes

The results of terrorist content review. Valid values:

normal
bloody
explosion
outfit
logo
weapon
politics
violence
crowd
parade
carcrash
flag
location
others

Count

Integer

Yes

The number of video snapshots.

TerrorismTopList

Field	Type	Required	Description
Label	String	Yes	The results of terrorist content review. Valid values: normal bloody explosion outfit logo weapon politics violence crowd parade carcrash flag location others
Score	String	Yes	The score of the video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Timestamp	String	Yes	The timestamp of the snapshot in the video. Unit: milliseconds.
Url	String	Yes	The URL of the video snapshot.

PornResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for pornographic content review results. Valid values: block review pass
Label	String	Yes	The results of pornographic content review. Valid values: porn: pornographic content sexy: lewd or obscene content normal
MaxScore	String	Yes	The highest score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
AverageScore	String	Yes	The average score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
CounterList	PornCounterList[]	Yes	The categories of the pornographic content review results and the number of video snapshots in each category.
TopList	PornTopList[]	Yes	The information about the top 10 video snapshots that have the highest scores in the category.

PornCounterList

Field

Type

Required

Description

Label

String

Yes

The results of pornographic content review. Valid values:

porn: pornographic content
sexy: lewd or obscene content
normal

Count

Integer

Yes

The number of video snapshots.

PornTopList

Field	Type	Required	Description
Label	String	Yes	The results of pornographic content review. Valid values: porn: pornographic content sexy: lewd or obscene content normal
Score	String	Yes	The score of the video snapshot that hits the label. Valid values: [0, 100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Timestamp	String	Yes	The timestamp of the snapshot in the video. Unit: milliseconds.
Url	String	Yes	The URL of the video snapshot.

AdResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for ad review results. Valid values: block review pass
Label	String	Yes	The results of ad review. Valid values: normal ad: ad violation politics: politically sensitive content in text porn: pornographic content in text abuse: abuse in text terrorism: terrorist content in text contraband: prohibited content in text spam: junk content in text npx: illegal ad qrcode: QR code programCode: mini program code
MaxScore	String	Yes	The highest score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
AverageScore	String	Yes	The average score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
CounterList	AdCounterList[]	Yes	The categories of the ad review results and the number of video snapshots in each category.
TopList	AdTopList[]	Yes	The information about the top 10 video snapshots that have the highest scores in the category.

AdCounterList

Field

Type

Required

Description

Label

String

Yes

The categories of the ad review results. Valid values:

normal
ad: ad violation
politics: politically sensitive content in text
porn: pornographic content in text
abuse: abuse in text
terrorism: terrorist content in text
contraband: prohibited content in text
spam: junk content in text
npx: illegal ad
qrcode: QR code
programCode: mini program code

Count

Integer

Yes

The number of video snapshots.

AdTopList

Field	Type	Required	Description
Label	String	Yes	The categories of the ad review results. Valid values: normal ad: ad violation politics: politically sensitive content in text porn: pornographic content in text abuse: abuse in text terrorism: terrorist content in text contraband: prohibited content in text spam: junk content in text npx: illegal ad qrcode: QR code programCode: mini program code
Score	String	Yes	The score of the video snapshot that hits the label. Valid values: [0, 100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Timestamp	String	Yes	The timestamp of the snapshot in the video. Unit: milliseconds.
Url	String	Yes	The URL of the video snapshot.

LiveResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for undesired content review results. Valid values: block review pass
Label	String	Yes	The results of undesired content review. Valid values: normal meaningless: no content in the image, such as black or white screen PIP: picture-in-picture smoking drivelive: live streaming in a running vehicle
MaxScore	String	Yes	The highest score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
AverageScore	String	Yes	The average score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
CounterList	LiveCounterList[]	Yes	The categories of the undesired content review results and the number of video snapshots in each category.
TopList	LiveTopList[]	Yes	The information about the top 10 video snapshots that have the highest scores in the category.

LiveCounterList

Field

Type

Required

Description

Label

String

Yes

The categories of undesired content review results. Valid values:

normal
meaningless: no content in the image, such as black or white screen
PIP: picture-in-picture
smoking
drivelive: live streaming in a running vehicle

Count

Integer

Yes

The number of video snapshots.

LiveTopList

Field	Type	Required	Description
Label	String	Yes	The categories of undesired content review results. Valid values: normal meaningless: no content in the image, such as black or white screen PIP: picture-in-picture smoking drivelive: live streaming in a running vehicle
Score	String	Yes	The score of the video snapshot that hits the label. Valid values: [0, 100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Timestamp	String	Yes	The timestamp of the snapshot in the video. Unit: milliseconds.
Url	String	Yes	The URL of the video snapshot.

LogoResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for logo review results. Valid values: block review pass
Label	String	Yes	The results of logo review. Valid values: normal TV: controlled logo trademark
MaxScore	String	Yes	The highest score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
AverageScore	String	Yes	The average score of a video snapshot that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
CounterList	LogoCounterList[]	Yes	The categories of the logo review results and the number of video snapshots in each category.
TopList	LogoTopList[]	Yes	The information about the top 10 video snapshots that have the highest scores in the category.

LogoCounterList

Field

Type

Required

Description

Label

String

Yes

The categories of logo review results. Valid values:

normal
TV: controlled logo
trademark

Count

Integer

Yes

The number of video snapshots.

LogoTopList

Field	Type	Required	Description
Label	String	Yes	The categories of logo review results. Valid values: normal TV: controlled logo trademark
Score	String	Yes	The score of the video snapshot that hits the label. Valid values: [0, 100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Timestamp	String	Yes	The timestamp of the snapshot in the video. Unit: milliseconds.
Url	String	Yes	The URL of the video snapshot.

CoverResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for the video thumbnail review results. Valid values: block review pass
Label	String	Yes	The categories of the video thumbnail review results. Separate multiple values with commas (,). Valid values: porn: pornographic content terrorism: terrorist or politically sensitive content ad: ad violation live: undesirable scene logo: logo image normal
Type	String	Yes	The reviewed content. Valid value: cover.
Url	String	Yes	The URL of the video thumbnail.
Result	CoverResultItem[]	Yes	The review results of the video thumbnail.

CoverResultItem

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for the video thumbnail review results. Valid values: block review pass
Label	String	Yes	The categories of the video thumbnail review results. Valid values if Scene is porn: porn: pornographic content sexy: lewd or obscene content normal Valid values if Scene is terrorism: normal bloody explosion outfit logo weapon politics violence crowd parade carcrash flag location others Valid values if Scene is ad: normal ad: ad violation politics: politically sensitive content in text porn: pornographic content in text abuse: abuse in text terrorism: terrorist content in text contraband: prohibited content in text spam: junk content in text npx: illegal ad qrcode: QR code programCode: mini program code Valid values if Scene is live: normal meaningless: no content in the image, such as black or white screen PIP: picture-in-picture smoking drivelive: live streaming in a running vehicle Valid values if Scene is logo: normal TV: controlled logo trademark
Score	String	Yes	The score of the video thumbnail that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Scene	String	Yes	The review type for the video thumbnail. Valid values: porn: pornographic content terrorism: terrorist or politically sensitive content ad: ad violation live: undesirable scene logo: logo image

TitleResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for text review results. Valid values: block review pass
Label	String	Yes	The categories of the text review results. Valid values: spam ad: ad violation abuse: abusive content flood: excessive junk content contraband: prohibited content meaningless: meaningless content normal
Score	String	Yes	The score of the image that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Scene	String	Yes	The review type. Valid value: antispam.
Type	String	Yes	The type of the text. Valid value: title.

AudioResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for audio review results. Valid values: block review pass
Label	String	Yes	The categories of the audio review results. Valid values: normal spam ad: ad violation politics terrorism: terrorist content abuse: abusive content porn: pornographic content flood: excessive junk content contraband: prohibited content meaningless: meaningless content
Score	String	Yes	The score of the audio that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Scene	String	Yes	The review type. Valid value: antispam.

ImageResult

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for image review results. Valid values: block review pass
Label	String	Yes	The categories of the image review results. Separate multiple values with commas (,). Valid values: porn: pornographic content terrorism: terrorist or politically sensitive content ad: ad violation live: undesirable scene logo: logo image normal
Url	String	Yes	The URL of the image.
Result	ImageResultItem[]	Yes	The results of image review.

ImageResultItem

Field	Type	Required	Description
Suggestion	String	Yes	The recommendation for image review results. Valid values: block review pass
Label	String	Yes	The categories of the image review results. Valid values if Scene is porn: porn: pornographic content sexy: lewd or obscene content normal Valid values if Scene is terrorism: normal bloody explosion outfit logo weapon politics violence crowd parade carcrash flag location others Valid values if Scene is ad: normal ad: ad violation politics: politically sensitive content in text porn: pornographic content in text abuse: abuse in text terrorism: terrorist content in text contraband: prohibited content in text spam: junk content in text npx: illegal ad qrcode: QR code programCode: mini program code Valid values if Scene is live: normal meaningless: no content in the image, such as black or white screen PIP: picture-in-picture smoking drivelive: live streaming in a running vehicle Valid values if Scene is logo: normal TV: controlled logo trademark
Score	String	Yes	The score of the image that hits the label. Valid values: [0,100]. The value is accurate to 10 decimal places. The score indicates the probability that the video snapshot is classified with the correct label. A higher value indicates a higher hit accuracy.
Scene	String	Yes	The review type for the image. Valid values: porn: pornographic content terrorism: terrorist or politically sensitive content ad: ad violation live: undesirable scene logo: logo image

AIMediaDNAResult

Parameter	Type	Required	Description
VideoDNA	VideoDNA[]	No	The video fingerprint recognition results.

VideoDNA

Parameter	Type	Required	Description
PrimaryKey	String	Yes	The ID of the video that has a similar fingerprint.
Similarity	String	Yes	The similarity between the fingerprints of the input video and the matched video.
Detail	VideoDNADetail[]	Yes	The details about similar videos, including the video location and duration.

VideoDNADetail

Parameter	Type	Required	Description
Input	VideoDNADetailItem	Yes	The start time and duration of the input video.
Duplication	VideoDNADetailItem	Yes	The start time and duration of the matched video in the media fingerprint repository.

VideoDNADetailItem

Parameter	Type	Required	Description
Start	String	Yes	The start time of the video.
Duration	String	Yes	The duration of the video.

AIVideoTagResult

Field	Type	Required	Description
Person	AIVideoTagResultItem[]	No	The figure tags.
Time	AIVideoTagResultItem[]	No	The time tags.
Location	AIVideoTagResultItem[]	No	The location tags.
Keyword	AIVideoTagResultItem[]	No	The keyword tags.
Category	AIVideoTagResultItem[]	No	The video categories.
OrigASRData	AIASRData[]	No	The original results of automated speech recognition (ASR). This field is returned only when the values of AnalyseTypes include ASR and OrigASRData is set to true. AnalyseTypes and OrigASRData are included in the request parameter Config that you must specify when you call SubmitAIJob.

AIVideoTagResultItem

Field	Type	Required	Description
Tag	String	Yes	The video tags.
Times	String[]	No	The time when the tag is displayed. Unit: milliseconds.
FaceUrl	String	No	The URL of the profile photo. This parameter is returned only when a figure tag was used.

AIASRData

Field	Type	Required	Description
Duration	String	Yes	The duration of the audio. Unit: milliseconds.
AsrTextList	AsrText[]	No	The results of speech recognition. Each element in the array is a sentence. For more information, see AsrText.

AsrText

Field	Type	Required	Description
StartTime	Long	Yes	The start time offset of the sentence. Unit: milliseconds.
EndTime	Long	Yes	The end time offset of the sentence. Unit: milliseconds.
Text	String	Yes	The recognition results of the sentence.
ChannelId	Int	Yes	The sound channel number.
SpeechRate	Int	Yes	The average speech rate of the sentence. Unit: words per minute.