Unpaired Translation of Point Clouds for Modeling Detector Response

TL;DR

This paper introduces a novel unpaired point cloud translation framework based on diffusion probabilistic models to improve detector response modeling in time projection chambers, aiding noise reduction and simulator accuracy.

Contribution

It presents a new diffusion-based method for unpaired point cloud translation, specifically applied to detector response modeling in physics experiments.

Findings

Successful translation between simulated and experimental data

Improved noise rejection in detector data

Enhanced fidelity of simulation models

Abstract

Modeling detector response is a key challenge in time projection chambers. We cast this problem as an unpaired point cloud translation task, between data collected from simulations and from experimental runs. Effective translation can assist with both noise rejection and the construction of high-fidelity simulators. Building on recent work in diffusion probabilistic models, we present a novel framework for performing this mapping. We demonstrate the success of our approach in both synthetic domains and in data sourced from the Active-Target Time Projection Chamber.