---
title: "Discussion on \"7 Things I Read in an AR Makeup SDK's Docs Before I Test It\" | Hashnode"
description: "Discussion on \"7 Things I Read in an AR Makeup SDK's Docs Before I Test It\". Segmentation granularity, makeup type counts, on-device vs server processing, download weight, wrapper ownership and billing shape."
url: "https://hashnode.com/posts/ar-makeup-sdk-comparison/6a992fbac9b172de09a0d815"
---

[Feed](https://hashnode.com/feed) /

Discussion

# [7 Things I Read in an AR Makeup SDK's Docs Before I Test It](https://dave-talks.hashnode.dev/ar-makeup-sdk-comparison?utm_source=hashnode&utm_medium=feed)

Segmentation granularity, makeup type counts, on-device vs server processing, download weight, wrapper ownership and billing shape.

Sep 3 · 16 min read · dave-talks.hashnode.dev  
Read the article

-   [#augmented-reality](https://hashnode.com/tag/augmented-reality)
-   [#computer-vision](https://hashnode.com/tag/computer-vision)
-   [#javascript](https://hashnode.com/tag/javascript)
-   [#webgl](https://hashnode.com/tag/webgl)
-   [#webdev](https://hashnode.com/tag/webdev)

## Written by

[Dave Gordon](https://hashnode.com/@davegordon)

@davegordon · Austin, USA

  
[View profile](https://hashnode.com/@davegordon)

Independent benchmarks. AI, AR, ML.

I build AR and ML features for mobile and web, and I test the SDKs that promise to make that easy. Most of them look great in a demo and very different on a five-year-old Android. I write down which parts hold up.

## Responses

Markdown supported · ⌘↵ to post

No responses yet.