Portrait Eyeglasses and Shadow Removal by Leveraging 3D Synthetic Data
Junfeng Lyu, Zhibo Wang, Feng Xu

TL;DR
This paper introduces a novel framework for removing eyeglasses and their cast shadows from portrait images using synthetic data and a detect-then-remove approach, improving face analysis tasks.
Contribution
It presents the first method to simultaneously remove eyeglasses and cast shadows using a synthetic dataset and cross-domain adaptation techniques.
Findings
Effective removal of eyeglasses and shadows demonstrated
Synthetic dataset improves training without real paired data
Cross-domain technique bridges synthetic and real data gap
Abstract
In portraits, eyeglasses may occlude facial regions and generate cast shadows on faces, which degrades the performance of many techniques like face verification and expression recognition. Portrait eyeglasses removal is critical in handling these problems. However, completely removing the eyeglasses is challenging because the lighting effects (e.g., cast shadows) caused by them are often complex. In this paper, we propose a novel framework to remove eyeglasses as well as their cast shadows from face images. The method works in a detect-then-remove manner, in which eyeglasses and cast shadows are both detected and then removed from images. Due to the lack of paired data for supervised training, we present a new synthetic portrait dataset with both intermediate and final supervisions for both the detection and removal tasks. Furthermore, we apply a cross-domain technique to fill the gap…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Code & Models
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsFace recognition and analysis · Facial Nerve Paralysis Treatment and Research · Facial Rejuvenation and Surgery Techniques
