VideoContentSearchOnScreenTextClusterFeature

AI Overview😉

  • The potential purpose of this module is to analyze and extract features from on-screen text in videos, such as OCR (Optical Character Recognition) text, ASR (Automatic Speech Recognition) text, and other metadata. This module aims to provide a comprehensive understanding of the text content within a video.
  • This module could impact search results by allowing Google to better understand the content of videos and provide more accurate and relevant search results. For example, if a user searches for a specific topic, Google can use this module to identify videos that contain relevant on-screen text, even if the text is not spoken or mentioned in the video's title or description.
  • To be more favorable for this function, a website or video content creator could ensure that their videos contain clear and concise on-screen text, such as captions, subtitles, or other metadata. This would allow the module to more accurately extract and analyze the text content, leading to better search result rankings. Additionally, providing high-quality video content with relevant and descriptive on-screen text could improve the overall user experience and increase engagement.

Interesting Module? Vote 👇

Voting helps other researchers find interesting modules.

Current Votes: 0

GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchOnScreenTextClusterFeature (google_api_content_warehouse v0.4.0)

Features for the set of OnScreenText.

Attributes

  • averageConfidence (type: number(), default: nil) - Average confidence.
  • averageDurationRatio (type: number(), default: nil) -
  • averageHorizontalPosition (type: number(), default: nil) -
  • averageOcrTextLength (type: number(), default: nil) - Deprecated. Please use ocr_text_length_stats.median instead. The average length of anchor labels. average_ocr_text_length is deprecated, because now ocr_text_length_stats has a field for holding the same value.
  • averageOfAverageTextHeightRatio (type: number(), default: nil) - Deprecated. Please use text_height_ratio_stats.mean instead. Average value of text height ratio (over image height), which is taken average over the same text. average_of_average_text_height_ratio is deprecated, because now text_height_ratio_stats has a field for keeping the same value.
  • averageVerticalPosition (type: number(), default: nil) - Average of label center position.
  • clusterRatio (type: number(), default: nil) - The number of anchors in the cluster over the number of anchors in the video.
  • clusterSize (type: integer(), default: nil) - The number of anchors in the cluster.
  • countingNumberRatio (type: number(), default: nil) - The number of anchors that had a counting number over the number of anchors in total.
  • durationMsStats (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t, default: nil) - Median, average and standard deviation of duration_ms among anchors in the same cluster.
  • frameSizeRatio (type: number(), default: nil) - Frame size ratio over total frames in video.
  • logOcrTextLengthStats (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t, default: nil) - Average and standard deviation of logarithm of the length of labels among anchors in the same cluster.
  • logTextHeightRatioStats (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t, default: nil) - Average and standard deviation of log(average_text_height_ratio) among anchors in the same cluster.
  • logp1000DurationMsStats (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t, default: nil) - Average and standard deviation of log(1000 + duration_ms) among anchors in the same cluster. Since duration_ms can be zero, 1000 is added before applying logarithm.
  • maxVideoDurationRatioBetweenAnchors (type: number(), default: nil) - The maximum ratio of duration between two consecutive anchors to video duration. This is calculated after all anchor filtering has been completed.
  • maximumDurationRatio (type: number(), default: nil) - Stats for ratio of frame time intervals, over total video time.
  • medianClusteringDistance (type: number(), default: nil) - The median cluster distance for the anchors in the cluster. The way the distance is calculated will vary depending on the clustering method.
  • medianDurationRatio (type: number(), default: nil) -
  • medianOfAverageTextHeightRatio (type: number(), default: nil) - Deprecated. Please use text_height_ratio_stats.median instead. Median value of text height ratio (over image height), which is taken average over the same text. median_of_average_text_height_ratio is deprecated, because now text_height_ratio_stats has a field for keeping the same value.
  • ocrAsrFeature (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchOcrAsrSetFeature.t, default: nil) - Features for the overlap between OCR and ASR.
  • ocrTextLengthStats (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t, default: nil) - Median, average and standard deviation of the length of labels among anchors in the same cluster.
  • stddevDurationRatio (type: number(), default: nil) -
  • textHeightRatioStats (type: GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t, default: nil) - Median, average and standard deviation of average_text_height_ratio among anchors in the same cluster.

Summary

Types

t()

Functions

decode(value, options)

Unwrap a decoded JSON object into its complex fields.

Types

Link to this type

t()

@type t() ::
  %GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchOnScreenTextClusterFeature{
    averageConfidence: number() | nil,
    averageDurationRatio: number() | nil,
    averageHorizontalPosition: number() | nil,
    averageOcrTextLength: number() | nil,
    averageOfAverageTextHeightRatio: number() | nil,
    averageVerticalPosition: number() | nil,
    clusterRatio: number() | nil,
    clusterSize: integer() | nil,
    countingNumberRatio: number() | nil,
    durationMsStats:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t()
      | nil,
    frameSizeRatio: number() | nil,
    logOcrTextLengthStats:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t()
      | nil,
    logTextHeightRatioStats:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t()
      | nil,
    logp1000DurationMsStats:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t()
      | nil,
    maxVideoDurationRatioBetweenAnchors: number() | nil,
    maximumDurationRatio: number() | nil,
    medianClusteringDistance: number() | nil,
    medianDurationRatio: number() | nil,
    medianOfAverageTextHeightRatio: number() | nil,
    ocrAsrFeature:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchOcrAsrSetFeature.t()
      | nil,
    ocrTextLengthStats:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t()
      | nil,
    stddevDurationRatio: number() | nil,
    textHeightRatioStats:
      GoogleApi.ContentWarehouse.V1.Model.VideoContentSearchMetricStats.t()
      | nil
  }

Functions

Link to this function

decode(value, options)

@spec decode(struct(), keyword()) :: struct()

Unwrap a decoded JSON object into its complex fields.