Skip to content
Yolact for iOS implemented using CoreML.
Swift
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
Yolact-CoreML Add mask display Sep 23, 2019
README.md
yolact.png

README.md

Yolact-CoreML

An example of You Only Look At CoefficienTs on iOS using CoreML.

About Yolact

Yolact (https://github.com/dbolya/yolact) is a simple, fully convolutional model for real-time instance segmentation. https://arxiv.org/abs/1904.02689

Yolact

Code references

Environment

onnx 1.4.1

onnx-coreml 0.4.0

onnx-simplifier 0.1.8

torch 1.0.1

Usage

How to convert pytorch model to CoreML model:

  1. Run yolcat code with ONNX and CoreML converter to convert to ONNX model (WITHOUT priors layer).

  2. Use onnx-simplifier to simplify ONNX model.

  3. Disable Upsample layer in onnx-coreml package to make Upsample custom layer in CoreML.

  4. Run ONNXToCoreML converter to conver ONNX mode to CoreML model, input should be MLMultiArray (3x550x550) and we need to do input normalization in our application code.

Todo

  1. Simply model to get more performance on mobile devices.
You can’t perform that action at this time.