NatML
Search…
MLVideoFeature
class NatML.Features.MLVideoFeature : MLFeature, IMLCloudFeature, IEnumerable<(MLImageFeature feature, long timestamp)>
INCOMPLETE.

Creating the Feature

INCOMPLETE.

From a Video File

/// <summary>
/// Create an video feature from a video file.
/// </summary>
/// <param name="path">Video file path.</param>
MLVideoFeature (string path);
INCOMPLETE.

From Streaming Assets

/// <summary>
/// Create a video feature from a video file in the `StreamingAssets` folder.
/// </summary>
/// <param name="relativePath">Relative path to video file in `StreamingAssets` folder.</param>
/// <returns>Video feature or `null` if no valid video can be found at the relative path.</returns>
static Task<MLVideoFeature> FromStreamingAssets (string relativePath);
INCOMPLETE.

Inspecting the Feature

INCOMPLETE

Feature Type

/// <summary>
/// Feature type.
/// </summary>
MLFeatureType type { get; }
Refer to the Inspecting the Feature section of the MLFeature class for more information.
The type is always an MLVideoType.

Video Path

/// <summary>
/// Video path.
/// </summary>
string path { get; }
The video feature provides the full path of its source video.
The video path always refers to an accessible file on the local file system.

Video Width

/// <summary>
/// Video width.
/// </summary>
int width { get; }
The video feature provides the width of its source video in pixels.

Video Height

/// <summary>
/// Video height.
/// </summary>
int height { get; }
The video feature provides the height of its source video in pixels.

Video Frame Count

/// <summary>
/// Video frame count.
/// </summary>
int frames { get; }
The video feature provides the frame count of its source video.
Note that the reported frame count is merely an approximate count based on the average frame rate and duration of the source video. Enumerate the video to get an exact count.

Handling Normalization

/// <summary>
/// Normalization mean.
/// </summary>
Vector4 mean { get; set; }
/// <summary>
/// Normalization standard deviation.
/// </summary>
Vector4 std { get; set }
See the Handling Normalization section of the MLImageFeature class for more information.

Handling Aspect Mode

/// <summary>
/// Aspect mode.
/// </summary>
MLImageFeature.AspectMode aspectMode { get; set; }
See the Handling Aspect Mode section of the MLImageFeature class for more information.

Enumerating Video Frames

INCOMPLETE.

Creating Cloud Features

/// <summary>
/// Create a Cloud ML feature that is ready for prediction with Cloud ML models.
/// </summary>
/// <param name="featureType">Optional feature type used to create the Cloud ML feature.</param>
/// <returns>Cloud ML feature.</returns>
MLCloudFeature IMLCloudFeature.Create (in MLFeatureType featureType = default);
INCOMPLETE.
Copy link
On this page
Creating the Feature
From a Video File
From Streaming Assets
Inspecting the Feature
Feature Type
Video Path
Video Width
Video Height
Video Frame Count
Handling Normalization
Handling Aspect Mode
Enumerating Video Frames
Creating Cloud Features