mlboydaisuke/3DDFA-V2-CoreML

3DDFA V2 — Core ML

3D Face Reconstruction, 2020

Single-image 3D face reconstruction. Predicts 6 DoF pose + expression parameters.

Core ML conversion of cleardusk/3DDFA_V2 for on-device inference on iPhone, iPad and Mac. Converted with coremltools; the packages are stateless, so all sequencing and buffering lives in your Swift code.

Task	keypoint detection
Upstream	cleardusk/3DDFA_V2
Packages	1
Download size	6 MB
Minimum iOS	17.0
Peak RAM	~200 MB

Files

File	Size	Compute units	SHA-256
3DDFA_V2.mlpackage.zip	6 MB	all	0f715dc220c046f5…
Total	6 MB

compute_units is not a suggestion -- it is the configuration the conversion was verified against. Moving a package to a different compute unit can silently change the numerics (FP16 attention overflow) or crash on the GPU.

Download

hf download mlboydaisuke/coreml-zoo --include "face3d/*" --local-dir ./face3d
unzip './face3d/face3d/*.zip' -d ./face3d

Use in Swift

import CoreML

let config = MLModelConfiguration()
config.computeUnits = .all   // as converted — see the table above

// Unzip the .mlpackage, drop it into your Xcode target and Xcode compiles it
// at build time:
let model = try 3DDFA_V2(configuration: config)

// ...or compile a downloaded .mlpackage at runtime:
let compiled = try await MLModel.compileModel(at: mlpackageURL)
let model = try MLModel(contentsOf: compiled, configuration: config)

Demo

• Sample app — sample_apps/Face3DDemo, a standalone SwiftUI project.
• Models Zoo — this model is downloadable and runnable inside the Models Zoo app on the App Store, no build required.

Conversion

• Pitfalls hit during conversion (FP16 overflow, ANE buffer limits, stride handling): docs/coreml_conversion_notes.md
• Model index: CoreML-Models

License

The conversion inherits the upstream license: MIT.

Credits

• Upstream authors: cleardusk/3DDFA_V2, 2020
• Core ML conversion: john-rocky (Daisuke Majima)