# SINet: Extreme Lightweight Portrait Segmentation Networks with Spatial Squeeze Modules and Information Blocking Decoder

arXiv: https://arxiv.org/abs/1911.09099 · HTML: https://mlsanity.com/paper/1911.09099/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Nov 2019.

## Built on by (most cited first, 5 of 5)

- [L-DAWA: Layer-wise Divergence Aware Weight Aggregation in Federated Self-Supervised Visual Representation Learning](https://mlsanity.com/paper/2307.07393/) (2023)
- [Federated Self-supervised Learning for Video Understanding](https://mlsanity.com/paper/2207.01975/) (2022)
- [Towards Label-Efficient Human Matting: A Simple Baseline for Weakly Semi-Supervised Trimap-Free Human Matting](https://mlsanity.com/paper/2404.00921/) (2024)
- [EasyPortrait -- Face Parsing and Portrait Segmentation Dataset](https://mlsanity.com/paper/2304.13509/) (2023)
- [Towards Imbalanced Motion: Part-Decoupling Network for Video Portrait Segmentation](https://mlsanity.com/paper/2307.16565/) (2023)

## Abstract (arXiv)

Designing a lightweight and robust portrait segmentation algorithm is an important task for a wide range of face applications. However, the problem has been considered as a subset of the object segmentation problem and less handled in the semantic segmentation field. Obviously, portrait segmentation has its unique requirements. First, because the portrait segmentation is performed in the middle of a whole process of many real-world applications, it requires extremely lightweight models. Second, there has not been any public datasets in this domain that contain a sufficient number of images with unbiased statistics. To solve the first problem, we introduce the new extremely lightweight portrait segmentation model SINet, containing an information blocking decoder and spatial squeeze modules. The information blocking decoder uses confidence estimates to recover local spatial information without spoiling global consistency. The spatial squeeze module uses multiple receptive fields to cope with various sizes of consistency in the image. To tackle the second problem, we propose a simple method to create additional portrait segmentation data which can improve accuracy on the EG1800 dataset. In our qualitative and quantitative analysis on the EG1800 dataset, we show that our method outperforms various existing lightweight segmentation models. Our method reduces the number of parameters from 2.1M to 86.9K (around 95.9% reduction), while maintaining the accuracy under an 1% margin from the state-of-the-art portrait segmentation method. We also show our model is successfully executed on a real mobile device with 100.6 FPS. In addition, we demonstrate that our method can be used for general semantic segmentation on the Cityscapes dataset. The code and dataset are available in https://github.com/HYOJINPARK/ExtPortraitSeg .