Published on Thu Aug 08 2019

ExtremeC3Net: Extreme Lightweight Portrait Segmentation Networks using Advanced C3-modules

Hyojin Park, Lars Lowe Sjösund, YoungJoon Yoo, Jihwan Bang, Nojun Kwak

Portrait segmentation is a subset of the object segmentation problem. The method reduces the number of parameters from 2.1M to 37.7K. It maintains the accuracy within a 1% margin from the current state-of-the-art method.

0
0
0
Abstract

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. bviously, portrait segmentation has its unique requirements. First, because the portrait segmentation is performed in the middle of a whole process of many realworld 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 problems, we introduce a new extremely lightweight portrait segmentation model consisting of a two-branched architecture based on the concentrated-comprehensive convolutions block. Our method reduces the number of parameters from 2.1M to 37.7K (around 98.2% reduction), while maintaining the accuracy within a 1% margin from the state-of-the-art portrait segmentation method. In our qualitative and quantitative analysis on the EG1800 dataset, we show that our method outperforms various existing lightweight segmentation models. Second, we propose a simple method to create additional portrait segmentation data which can improve accuracy on the EG1800 dataset. Also, we analyze the bias in public datasets by additionally annotating race, gender, and age on our own. The augmented dataset, the additional annotations and code are available in https://github.com/HYOJINPARK/ExtPortraitSeg .

Wed Nov 20 2019
Computer Vision
SINet: Extreme Lightweight Portrait Segmentation Networks with Spatial Squeeze Modules and Information Blocking Decoder
Portrait segmentation is performed in the middle of a whole process of many real-world applications. It requires extremely lightweight models. We introduce the new extremely lightweight portrait segmentation model SINet.
0
0
0
Sat Apr 22 2017
Computer Vision
On Face Segmentation, Face Swapping, and Face Perception
We show that even when face images are unconstrained and arbitrarily paired, face swapping between them is actually quite simple. We show that better face swapping produces less recognizable inter-subject results. This is the first time this effect was definitively demonstrated for machine vision systems.
0
0
0
Thu Dec 12 2019
Computer Vision
Zooming into Face Forensics: A Pixel-level Analysis
Face forensics techniques are needed to distinguish those who have been tampered with. A large scale dataset "FaceForensics++" has provided enormous training data. Previous works have focused on casting it as a global prediction problem.
0
0
0
Mon Aug 14 2017
Computer Vision
SSH: Single Stage Headless Face Detector
The Single Stage Headless (SSH) face detector is 5X faster than the current state-of-the-art. Unlike two stage proposals, SSH detects faces in a single stage directly from the convolutional layers in a classification network.
0
0
0
Thu Jun 25 2020
Computer Vision
PropagationNet: Propagate Points to Curve to Learn Structure Information
0
0
0
Thu Nov 26 2020
Computer Vision
TinaFace: Strong but Simple Baseline for Face Detection
Face detection has received intensive attention in recent years. There is no gap between face detection and generic object detection. We provide a strong but simple baseline method to deal with face detection named TinaFace.
0
0
0