"It Is Hard to Remove from My Eye": Design Makeup Residue Visualization System for Chinese Traditional Opera (Xiqu) Performers

TL;DR

The paper addresses skin-health risks faced by Chinese traditional opera (Xiqu) performers from heavy-metal makeup and incomplete removal, especially around the eyes. It adopts a formative study (online survey N=136 and interviews N=15) to derive six design considerations and introduces EyeVis, a portable system that visualizes eye-makeup residue and records wearing time via a mobile app plus a camera-magnifier, fill light, and eye shields. EyeVis employs non-ML vision-based algorithms (eye feature localization, HSV-UV simulation, and binary threshold) and is evaluated through a seven-day deployment with 12 performers, demonstrating improved makeup removal completeness, reduced effort, and measurable residue reduction, along with robust illumination and user feedback. The work advances skincare tooling for performers, supports preservation of intangible cultural heritage, and suggests generalizations to broader skin areas and makeup contexts, including skincare data collection and wider audiences.

Abstract

Chinese traditional opera (Xiqu) performers often experience skin problems due to the long-term use of heavy-metal-laden face paints. To explore the current skincare challenges encountered by Xiqu performers, we conducted an online survey (N=136) and semi-structured interviews (N=15) as a formative study. We found that incomplete makeup removal is the leading cause of human-induced skin problems, especially the difficulty in removing eye makeup. Therefore, we proposed EyeVis, a prototype that can visualize the residual eye makeup and record the time make-up was worn by Xiqu performers. We conducted a 7-day deployment study (N=12) to evaluate EyeVis. Results indicate that EyeVis helps to increase Xiqu performers' awareness about removing makeup, as well as boosting their confidence and security in skincare. Overall, this work also provides implications for studying the work of people who wear makeup on a daily basis, and helps to promote and preserve the intangible cultural heritage of practitioners.

Figures (9)

• Figure 1: Structural Prototype Assemble: Camera Lens Magnifier (in cyan), Fill Light (in orange), and Eye Shields (in purple)
• Figure 2: Mobile App Workflow: (1) Take photos of eyes without makeup (whole face, open and close eyes), (2) Timing and visualization (HSV-UV simulation and Binary threshold, comparison w / o makeup), (3) Time trend and record display.
• Figure 3: Sample pipeline of Eye feature points localization. (1) Run Facemesh on whole face as a reference to find the bounding box of eye feature points, (2) Resize the origin image to the bounding box size and replace it with the eye area of the whole face, (3) Run Facemesh again with the input combined image, (4) Extract the eye feature points and resize back to the final result.
• Figure 4: Overview of Vision-based Algorithms: firstly run (A) eye feature points localization (clip and resize the image based on the bounding box), then run (B) HSV-UV simulation (we provide blue color for black face paint, and red color for pink face paint visualization) and (C) Binary Threshold to visualize the makeup residual.
• Figure 5: LEFT: Sample Illumination Test: w / o EyeVis in HSV color space. RIGHT: Actual lighting surrounding environments. The images are taken under 3 different lighting conditions: a) without room light, b) under room light, and c) under sunshine.