English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Designing a lightweight and robust portrait segmentation algorithm is animportant task for a wide range of face applications. However, the problem hasbeen considered as a subset of the object segmentation problem and less handledin the semantic segmentation field. Obviously, portrait segmentation has itsunique requirements. First, because the portrait segmentation is performed inthe middle of a whole process of many real-world applications, it requiresextremely lightweight models. Second, there has not been any public datasets inthis domain that contain a sufficient number of images with unbiasedstatistics. To solve the first problem, we introduce the new extremelylightweight portrait segmentation model SINet, containing an informationblocking decoder and spatial squeeze modules. The information blocking decoderuses confidence estimates to recover local spatial information without spoilingglobal consistency. The spatial squeeze module uses multiple receptive fieldsto cope with various sizes of consistency in the image. To tackle the secondproblem, we propose a simple method to create additional portrait segmentationdata which can improve accuracy on the EG1800 dataset. In our qualitative andquantitative analysis on the EG1800 dataset, we show that our methodoutperforms various existing lightweight segmentation models. Our methodreduces 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-artportrait segmentation method. We also show our model is successfully executedon a real mobile device with 100.6 FPS. In addition, we demonstrate that ourmethod can be used for general semantic segmentation on the Cityscapes dataset.The code and dataset are available inhttps://github.com/HYOJINPARK/ExtPortraitSeg .

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