PoreTrack3D: A Benchmark for Dynamic 3D Gaussian Splatting in Pore-Scale Facial Trajectory Tracking

TL;DR

PoreTrack3D introduces a comprehensive benchmark dataset for dynamic 3D Gaussian splatting in pore-scale facial trajectory tracking, enabling detailed analysis of subtle facial movements and establishing new performance baselines.

Contribution

This paper presents the first benchmark dataset for pore-scale facial trajectory tracking using 3D Gaussian splatting, including extensive data and evaluation protocols.

Findings

Established baseline performance for state-of-the-art methods.

Provided a large-scale dataset with over 440,000 trajectories.

Demonstrated the feasibility of high-fidelity facial motion capture.

Abstract

We introduce PoreTrack3D, the first benchmark for dynamic 3D Gaussian splatting in pore-scale, non-rigid 3D facial trajectory tracking. It contains over 440,000 facial trajectories in total, among which more than 52,000 are longer than 10 frames, including 68 manually reviewed trajectories that span the entire 150 frames. To the best of our knowledge, PoreTrack3D is the first benchmark dataset to capture both traditional facial landmarks and pore-scale keypoints trajectory, advancing the study of fine-grained facial expressions through the analysis of subtle skin-surface motion. We systematically evaluate state-of-the-art dynamic 3D Gaussian splatting methods on PoreTrack3D, establishing the first performance baseline in this domain. Overall, the pipeline developed for this benchmark dataset's creation establishes a new framework for high-fidelity facial motion capture and dynamic 3D reconstruction. Our dataset are publicly available at: https://github.com/JHXion9/PoreTrack3D