https://github.com/lukereichold/visualactionkit

Human action classification for video, offline and natively on iOS via Core ML
https://github.com/lukereichold/visualactionkit

3d-cnn avkit convolutional-neural-network coreml ios kinetics neural-network swift swiftui video-classification video-processing vision

Last synced: 4 months ago
JSON representation

Human action classification for video, offline and natively on iOS via Core ML

• Host: GitHub
• URL: https://github.com/lukereichold/visualactionkit
• Owner: lukereichold
• License: mit
• Created: 2020-07-21T05:39:21.000Z (almost 6 years ago)
• Default Branch: master
• Last Pushed: 2020-08-05T06:57:18.000Z (almost 6 years ago)
• Last Synced: 2025-11-20T17:07:43.763Z (7 months ago)
• Topics: 3d-cnn, avkit, convolutional-neural-network, coreml, ios, kinetics, neural-network, swift, swiftui, video-classification, video-processing, vision
• Language: Swift
• Homepage: https://lukereichold.com/blog/posts/video-action-classification-coreml/
• Size: 51.2 MB
• Stars: 24
• Watchers: 0
• Forks: 4
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# VisualActionKit

![](https://img.shields.io/badge/Swift-5.3-orange.svg)

[![GitHub license](https://img.shields.io/badge/license-MIT-lightgrey.svg)](https://github.com/lukereichold/VisualActionKit/blob/master/LICENSE) 

[![SPM compatible](https://img.shields.io/badge/spm-compatible-brightgreen.svg?style=flat)](https://swift.org/package-manager)

[![Twitter](https://img.shields.io/badge/twitter-@lreichold-blue.svg?style=flat)](https://twitter.com/lreichold)

Human action classification for video, offline and natively on iOS via Core ML

Uses the [`kinetics-i3d`]() model to classify videos into one of 400 different action classes defined in [Kinetics 400](https://deepmind.com/research/open-source/kinetics)

⚠️ _This project requires Xcode 12._

**Reference**: [See accompanying blog post](https://lukereichold.com/blog/posts/video-action-classification-coreml/)

[![Open in Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/drive/11Cr8w_T53bPLHPvLfZteAxHhtjgBS-LH)

## Installation

To install via [Swift Package Manager](https://swift.org/package-manager), add `VisualActionKit` to your `Package.swift` file. Alternatively, add it from Xcode directly.

```swift

let package = Package(

    ...

    dependencies: [

        .package(url: "https://github.com/lukereichold/VisualActionKit.git", from: "0.1.0")

    ],

    ...

)

```

Then import `VisualActionKit` wherever you’d like to use it:

```swift

import VisualActionKit

```

## Usage

```swift

let url = Bundle.module.url(forResource: "writing", withExtension: "mp4")

let asset = AVAsset(url: url)

try Classifier.shared.classify(asset) { predictions in

    print(predictions)

}

```

## Contribute

Contributions welcome. Please check out [the issues](https://github.com/lukereichold/VisualActionKit/issues).