SplatPose+: Real-time Image-Based Pose-Agnostic 3D Anomaly Detection

TL;DR

This work proposes SplatPose+, which employs a hybrid representation consisting of a Structure from Motion model for localization and a 3D Gaussian Splatting model for Novel View Synthesis and offers real-time inference speeds and faster training compared to SplatPose.

Abstract

Image-based Pose-Agnostic 3D Anomaly Detection is an important task that has emerged in industrial quality control. This task seeks to find anomalies from query images of a tested object given a set of reference images of an anomaly-free object. The challenge is that the query views (a.k.a poses) are unknown and can be different from the reference views. Currently, new methods such as OmniposeAD and SplatPose have emerged to bridge the gap by synthesizing pseudo reference images at the query views for pixel-to-pixel comparison. However, none of these methods can infer in real-time, which is critical in industrial quality control for massive production. For this reason, we propose SplatPose+, which employs a hybrid representation consisting of a Structure from Motion (SfM) model for localization and a 3D Gaussian Splatting (3DGS) model for Novel View Synthesis. Although our proposed pipeline requires the computation of an additional SfM model, it offers real-time inference speeds and faster training compared to SplatPose. Quality-wise, we achieved a new SOTA on the Pose-agnostic Anomaly Detection benchmark with the Multi-Pose Anomaly Detection (MAD-SIM) dataset.

Figures (3)

• Figure 1: SplatPose+ pipeline incorporates an SfM model for localizing query views and initialization for the 3DGS model. 3DGS model is trained only for Novel View Synthesis.
• Figure 2: Defect Sample in MAD Datasetzhou2023paddatasetbenchmarkposeagnostic
• Figure 3: Quantitative Comparison for Anomaly Detection Performance on Sparse-View Reference Images