Source author record

David Peter

David Peter appears in the imported research catalog. Authorship, coauthor and topic links are available while profile ownership is still unclaimed.

ResearcherUnclaimed source record

cond-mat.quant-gas Computer Vision Artificial Intelligence Machine Learning physics.atom-ph quant-ph

Catalog footprint

What is connected

8works

6topics

4close collaborators

Actions

Connect this record

Open graph Browse works

Inspect adjacent papers, topics, institutions and collaborators without losing the researcher page.

Building this map preview

BZPEER is loading the nearby papers, people, topics and institutions for this page.

preprint2022arXiv

A Realism Metric for Generated LiDAR Point Clouds

A considerable amount of research is concerned with the generation of realistic sensor data. LiDAR point clouds are generated by complex simulations or learned generative models. The generated data is usually exploited to enable or improve downstream perception algorithms. Two major questions arise from these procedures: First, how to evaluate the realism of the generated data? Second, does more realistic data also lead to better perception performance? This paper addresses both questions and presents a novel metric to quantify the realism of LiDAR point clouds. Relevant features are learned from real-world and synthetic point clouds by training on a proxy classification task. In a series of experiments, we demonstrate the application of our metric to determine the realism of generated LiDAR data and compare the realism estimation of our metric to the performance of a segmentation model. We confirm that our metric provides an indication for the downstream segmentation performance.

preprint2022arXiv

Point Cloud Generation with Continuous Conditioning

Generative models can be used to synthesize 3D objects of high quality and diversity. However, there is typically no control over the properties of the generated object.This paper proposes a novel generative adversarial network (GAN) setup that generates 3D point cloud shapes conditioned on a continuous parameter. In an exemplary application, we use this to guide the generative process to create a 3D object with a custom-fit shape. We formulate this generation process in a multi-task setting by using the concept of auxiliary classifier GANs. Further, we propose to sample the generator label input for training from a kernel density estimation (KDE) of the dataset. Our ablations show that this leads to significant performance increase in regions with few samples. Extensive quantitative and qualitative experiments show that we gain explicit control over the object dimensions while maintaining good generation quality and diversity.

preprint2021arXiv

Quantifying point cloud realism through adversarially learned latent representations

Judging the quality of samples synthesized by generative models can be tedious and time consuming, especially for complex data structures, such as point clouds. This paper presents a novel approach to quantify the realism of local regions in LiDAR point clouds. Relevant features are learned from real-world and synthetic point clouds by training on a proxy classification task. Inspired by fair networks, we use an adversarial technique to discourage the encoding of dataset-specific information. The resulting metric can assign a quality score to samples without requiring any task specific annotations. In a series of experiments, we confirm the soundness of our metric by applying it in controllable task setups and on unseen data. Additional experiments show reliable interpolation capabilities of the metric between data with varying degree of realism. As one important application, we demonstrate how the local realism score can be used for anomaly detection in point clouds.

preprint2015arXiv

Topological bands with Chern number C=2 by dipolar exchange interactions

We demonstrate the realization of topological band structures by exploiting the intrinsic spin-orbit coupling of dipolar interactions in combination with broken time-reversal symmetry. The system is based on polar molecules trapped in a deep optical lattice, where the dynamics of rotational excitations follows a hopping Hamiltonian which is determined by the dipolar exchange interactions. We find topological bands with Chern number $C=2$ on the square lattice, while a very rich structure of different topological bands appears on the honeycomb lattice. We show that the system is robust against missing molecules. For certain parameters we obtain flat bands, providing a promising candidate for the realization of hard-core bosonic fractional Chern insulators.

preprint2014arXiv

Ferroelectric quantum phase transition with cold polar molecules

We analyze a system of polar molecules in a one-dimensional optical lattice. By controlling the internal structure of the polar molecules with static electric and microwave fields, we demonstrate the appearance of a quantum phase transition into a ferroelectric phase via spontaneous breaking of a $U(1)$ symmetry. The phase diagram is first analyzed within mean-field theory, while in a second step the results are verified by a mapping onto the Bose-Hubbard model for hard-core bosons. The latter is studied within the well-established bosonization procedure. We find that the ferroelectric phase is characterized by (quasi) long-range order for the electric dipole moments.

preprint2013arXiv

Coupling a single electron to a Bose-Einstein condensate

The coupling of electrons to matter is at the heart of our understanding of material properties such as electrical conductivity. One of the most intriguing effects is that electron-phonon coupling can lead to the formation of a Cooper pair out of two repelling electrons, the basis for BCS superconductivity. Here we study the interaction of a single localized electron with a Bose-Einstein condensate (BEC) and show that it can excite phonons and eventually set the whole condensate into a collective oscillation. We find that the coupling is surprisingly strong as compared to ionic impurities due to the more favorable mass ratio. The electron is held in place by a single charged ionic core forming a Rydberg bound state. This Rydberg electron is described by a wavefunction extending to a size comparable to the dimensions of the BEC, namely up to 8 micrometers. In such a state, corresponding to a principal quantum number of n=202, the Rydberg electron is interacting with several tens of thousands of condensed atoms contained within its orbit. We observe surprisingly long lifetimes and finite size effects due to the electron exploring the wings of the BEC. Based on our results we anticipate future experiments on electron wavefunction imaging, investigation of phonon mediated coupling of single electrons, and applications in quantum optics.

preprint2012arXiv

Anomalous Behavior of Spin Systems with Dipolar Interactions

We study the properties of spin systems realized by cold polar molecules interacting via dipole-dipole interactions in two dimensions. Using a spin wave theory, that allows for the full treatment of the characteristic long-distance tail of the dipolar interaction, we find several anomalous features in the ground state correlations and the spin wave excitation spectrum, which are absent in their counterparts with short range interaction. The most striking consequence is the existence of true long-range order at finite temperature for a two-dimensional phase with a broken U(1) symmetry.