mediapipe/docs/solutions/models.md

layout	title	parent	nav_order
default	Models and Model Cards	Solutions	30

MediaPipe Models and Model Cards

{: .no_toc }

1. TOC {:toc}

---

Attention: Thank you for your interest in MediaPipe Solutions. We have ended support for these MediaPipe Legacy Solutions as of March 1, 2023. All other MediaPipe Legacy Solutions will be upgraded to a new MediaPipe Solution. The code repository and prebuilt binaries for all MediaPipe Legacy Solutions will continue to be provided on an as-is basis. We encourage you to check out the new MediaPipe Solutions at: https://developers.google.com/mediapipe/solutions

This notice and web page will be removed on April 3, 2023.

---

Face Detection

• Short-range model (best for faces within 2 meters from the camera): TFLite model, TFLite model quantized for EdgeTPU/Coral, Model card
• Full-range model (dense, best for faces within 5 meters from the camera): TFLite model, Model card
• Full-range model (sparse, best for faces within 5 meters from the camera): TFLite model, Model card

Full-range dense and sparse models have the same quality in terms of F-score however differ in underlying metrics. The dense model is slightly better in Recall whereas the sparse model outperforms the dense one in Precision. Speed-wise sparse model is ~30% faster when executing on CPU via XNNPACK whereas on GPU the models demonstrate comparable latencies. Depending on your application, you may prefer one over the other.

Face Mesh

• Face landmark model: TFLite model, TF.js model
• Face landmark model w/ attention (aka Attention Mesh): TFLite model
• Model card, Model card (w/ attention)

Iris

• Iris landmark model: TFLite model
• Model card

Hands

• Palm detection model: TFLite model (lite), TFLite model (full), TF.js model
• Hand landmark model: TFLite model (lite), TFLite model (full), TF.js model
• Model card

Pose

• Pose detection model: TFLite model
• Pose landmark model: TFLite model (lite), TFLite model (full), TFLite model (heavy)
• Model card

Holistic

• Hand recrop model: TFLite model

Selfie Segmentation

• TFLite model (general)
• TFLite model (landscape)
• Model card

Hair Segmentation

• TFLite model
• Model card

Object Detection

• TFLite model
• TFLite model quantized for EdgeTPU/Coral

Objectron

• TFLite model for shoes
• TFLite model for chairs
• TFLite model for cameras
• TFLite model for cups
• Single-stage TFLite model for shoes
• Single-stage TFLite model for chairs
• Model card

KNIFT

• TFLite model for up to 200 keypoints
• TFLite model for up to 400 keypoints
• TFLite model for up to 1000 keypoints
• Model card