logo
ResearchBunny Logo
Delving into High-Quality Synthetic Face Occlusion Segmentation Datasets

Computer Science

Delving into High-Quality Synthetic Face Occlusion Segmentation Datasets

K. T. R. Voo, L. Jiang, et al.

Discover the innovative research by Kenny T. R. Voo, Liming Jiang, and Chen Change Loy that delves into occlusion-aware face segmentation. The study introduces groundbreaking techniques and high-resolution datasets that are essential for advancing applications in computer vision.

00:00
00:00
Playback language: English
Abstract
This paper performs a comprehensive analysis of datasets for occlusion-aware face segmentation, a crucial task for many downstream applications. The authors propose two occlusion generation techniques: Naturalistic Occlusion Generation (NatOcc) for high-quality naturalistic synthetic occluded faces, and Random Occlusion Generation (RandOcc), a more general method. They introduce two high-resolution real-world occluded face datasets with fine-grained annotations, RealOcc and RealOcc-Wild, and conduct a comprehensive analysis on a new segmentation benchmark.
Publisher
Published On
Authors
Kenny T. R. Voo, Liming Jiang, Chen Change Loy
Tags
face segmentation
occlusion awareness
synthetic data
high-resolution datasets
computer vision
Listen, Learn & Level Up
Over 10,000 hours of research content in 25+ fields, available in 12+ languages.
No more digging through PDFs, just hit play and absorb the world's latest research in your language, on your time.
listen to research audio papers with researchbunny