SEMI-PointRend: Improved Semiconductor Wafer Defect Classification and Segmentation as Rendering

TL;DR

This paper introduces SEMI-PointRend, a novel application of the PointRend segmentation method to semiconductor wafer defect classification, achieving higher accuracy than traditional Mask R-CNN models.

Contribution

The paper presents SEMI-PointRend, integrating PointRend into defect segmentation, and demonstrates its superior performance over Mask R-CNN in wafer defect segmentation tasks.

Findings

SEMI-PointRend outperforms Mask R-CNN by up to 18.8% in segmentation mean average precision.

The method effectively generates high-resolution, precise defect segmentation masks.

PointRend can be flexibly integrated into existing segmentation architectures.

Abstract

In this study, we applied the PointRend (Point-based Rendering) method to semiconductor defect segmentation. PointRend is an iterative segmentation algorithm inspired by image rendering in computer graphics, a new image segmentation method that can generate high-resolution segmentation masks. It can also be flexibly integrated into common instance segmentation meta-architecture such as Mask-RCNN and semantic meta-architecture such as FCN. We implemented a model, termed as SEMI-PointRend, to generate precise segmentation masks by applying the PointRend neural network module. In this paper, we focus on comparing the defect segmentation predictions of SEMI-PointRend and Mask-RCNN for various defect types (line-collapse, single bridge, thin bridge, multi bridge non-horizontal). We show that SEMI-PointRend can outperforms Mask R-CNN by up to 18.8% in terms of segmentation mean average precision.