---
language:
  - en
thumbnail: "https://i.imgur.com/z0BwlOe.png"
tags:
- image matting
- image segmentation
license: apache-2.0
metrics:
- mean_iou
---
PP-HumanSeg v1 model, released by [Paddle](https://github.com/PaddlePaddle/PaddleSeg/tree/release/2.6/contrib/PP-HumanSeg).

Tested on the [PP-HumanSeg-14K](https://github.com/PaddlePaddle/PaddleSeg/blob/release/2.6/contrib/PP-HumanSeg/paper.md) dataset.

| Model Name | Best Input Shape | mIou(%) | Inference Time on Arm CPU(ms) | Modle Size(MB) |
| --- | --- | --- | ---| --- |
| PP-HumanSegV1-Lite | 398x224 | 93.60 | 29.68 | 2.3 |

**Network Architecture**

Model architecture named ConnectNet, which is suitable for real-time segmentation scenarios on the web or mobile.

![pphumanseg_lite.png](https://raw.githubusercontent.com/unography/PaddleSeg/release/2.6/configs/pp_humanseg_lite/pphumanseg_lite.png)

[More details](https://github.com/unography/PaddleSeg/tree/release/2.6/configs/pp_humanseg_lite)