Packages

  • package root
    Definition Classes
    root
  • package com
    Definition Classes
    root
  • package nvidia
    Definition Classes
    com
  • package mdx

    this is the base transform for mdx analytics SDK, the transform comprise of

    this is the base transform for mdx analytics SDK, the transform comprise of


    core : mdx streaming core classes
    xapp : comprise of sample applications

    Definition Classes
    nvidia
  • package core

    this is the core transform for mdx SDK, contains transform and classes for the core mdx functionality

    this is the core transform for mdx SDK, contains transform and classes for the core mdx functionality

    Definition Classes
    mdx
  • package schema

    please see nv.schema , which is used @since v1.0

    please see nv.schema , which is used @since v1.0

    this is kept for backward compatibility and is used for json based pipeline

    the key classes representing the json schema

    Calibration
    Message
    Frame and
    Behavior

    There are two schema for sending messages perception layer to Message broker (kafka or IOT hub), The two schema are represented by Message and Frame. Frame is concise in terms amount of bytes sent over the network

    User can configure to use either of the Messaging schema

    jBehavior is specific to the domain we are dealing with, represent object movement & state over period of times

    Definition Classes
    core
  • package calibration

    classes related to Calibration JSON

    classes related to Calibration JSON

    Definition Classes
    schema
  • package frame
    Definition Classes
    schema
  • jCluster
  • jFrame
  • jFrameMessage
  • jFrameWithOffset
  • jPoint
  • jPointWithId
  • jRoiCount
  • jTypeCount
  • package puck
    Definition Classes
    schema
  • package trajectory

    transform for Trajectory schema

    transform for Trajectory schema

    Definition Classes
    schema

package frame

Type Members

  1. case class jCluster(points: Array[jPoint]) extends Product with Serializable

    proximity detection cluster

  2. case class jFrame(version: String, id: String, timestamp: Timestamp, sensorId: String, objects: Array[String]) extends Product with Serializable

    this is alternative to com.nvidia.mdx.core.schema.jMessage

    this is alternative to com.nvidia.mdx.core.schema.jMessage

    Each frame Message / payload have a one to one correspondence with Frame (Video Frame). A frame comprises of one or more objects detected.

    The JSON structure of the frame is
    
    {
    "version": "4.0",
    "id": "frame-id",
    "@timestamp": "2018-04-11T04:59:59.828Z",
    "sensorId": "sensor-id",
    "objects": [
    ".......object-1 attributes...........",
    ".......object-2 attributes...........",
    ".......object-3 attributes..........."
    ]
    }
    
    Example frame with only Primary GIE
    
    
    {
    "version": "4.0",
    "id": "frame-id",
    "@timestamp": "2018-04-11T04:59:59.828Z",
    "sensorId": "sensor-id",
    "objects": [
    "object-Id | bbox.leftX | bbox.topY | bbox.rightX | bbox.bottomY |object-type",
    ".......object-2 PGIE attributes...........",
    ".......object-3 PGIE attributes..........."
    ]
    }
    
    
    An example with SGIE  where object-type is Vehicle
    
    {
    "version": "4.0",
    "id": "frame-id",
    "@timestamp": "2018-04-11T04:59:59.828Z",
    "sensorId": "sensor-id",
    "objects": [
    "957|1834|150|1918|215|Vehicle|#|sedan|Bugatti|M|blue|CA 444|California|0.8",
    "..........."
    
    ]
    }
    version

    represents version of the schema

    id

    represents the video frame-id

    timestamp

    represents the camera timestamp

    sensorId

    Unique sensor-id

    objects

    List of objects, where each object is represented using a string, the attributes of the object are pipe seperated

    Since

    v0.1

  3. case class jFrameMessage(version: String, id: String, timestamp: Timestamp, sensorId: String, objects: Array[String], objectCoordinates: Array[jPointWithId], objectCount: Int, clusterThreshold: Double, clusters: Array[jCluster], proximityDetection: Int, rois: Array[jTypeCount] = null, fov: Array[jTypeCount] = null) extends Product with Serializable

    enhanced version of com.nvidia.mdx.core.schema.frame.jFrame.

    enhanced version of com.nvidia.mdx.core.schema.frame.jFrame. Also have added information about proximity detection or social distancing violation contains cluster of objects which violates social distancing

    version

    represents version of the schema

    id

    represents the video frame-id

    timestamp

    represents the camera timestamp

    sensorId

    Unique sensor-id

    objects

    List of objects, where each object is represented using a string, the attributes of the object are pipe seperated

    objectCoordinates

    Array of Points representing bbox

    objectCount

    number of object in the frame

    clusterThreshold

    distance threshold in meters, used for proximity detection

    clusters

    clusters of objects violating social distancing,

    proximityDetection

    number of groups violating social distancing

    rois

    array of object-type and their counts

    fov

    array of object-type and their counts

    Since

    v0.2

  4. case class jFrameWithOffset(version: String, id: String, timestamp: Timestamp, offset: Long, sensorId: String, objects: Array[String]) extends Product with Serializable
  5. case class jPoint(x: Double, y: Double) extends Product with Serializable
  6. case class jPointWithId(id: String, type: String, coordinate: jPoint) extends Product with Serializable
  7. case class jRoiCount(id: String, count: Int) extends Product with Serializable
  8. case class jTypeCount(id: String, type: String, count: Int) extends Product with Serializable

Ungrouped