[ Tech stack ]
MediaPipe
Google's on-device ML solutions: hands, pose, face and gestures in real time.
MediaPipe ships ready-made tasks that run entirely on the device: hand, pose and face detection with landmarks, gesture recognition, image segmentation, and an LLM Inference API to run a compact language model locally. The same Tasks API is available in Kotlin on Android, Swift on iOS, JavaScript on the web and Python, with Model Maker to adapt a model to your own data. It's our choice when an app has to understand the user's body or face in real time without sending video to a server.
[ Why MediaPipe at Dexon ]
What this technology does well, and why we use it.
Typical usage: Virtual try-on, fitness coaching, gesture-controlled interfaces, live filters and effects.
- 01
Real-time hand, pose and face landmarks with on-device GPU acceleration.
- 02
One Tasks API: Android (Kotlin), iOS (Swift), web (JavaScript) and Python.
- 03
Ready-made gesture recognition, segmentation, classification and object detection.
- 04
LLM Inference API: a compact LLM such as Gemma running offline on the phone.
[ Complementary technologies ]
The building blocks we often mobilise alongside.
A stack rarely exists alone. Here are the technologies Dexon most often pairs with this one, through pipeline habits, usage similarity or internal mastery. Click on a building block to see its scope.
[ Reassurance ]
- 0+
- custom projects delivered
- 30
- engineers, designers, project managers
- 80 %
- from top French schools
- 24 h
- average reply time
[ They trust us ]
More than 100 French and European companies trust us
[ Press ]
They talk about us.
Nationwide coverage by BFM Business, Le Figaro, Challenges, La Tribune and CNews. An outside reading of our work and our innovations.
