arXiv subjects

Minjung Son

Publications and source records attributed to Minjung Son.

4 recordsLinked to original sources

arXiv2025

Two-dimensional electronic spectroscopy in the condensed phase using equivariant transformer accelerated molecular dynamics simulations

Two-dimensional electronic spectroscopy (2DES) provides rich information about how the electronic states of molecules, proteins, and solid-state materials interact with each other and their surrounding environment. Atomistic molecular dynamics simulations offer an appealing route to uncover how nuclear motions mediate electronic energy relaxation and their manifestation in electronic spectroscopies, but are computationally expensive. Here we show that, by using an equivariant transformer-based machine learning architecture trained with only ~2500 ground state and ~100 excited state electronic structure calculations, one can construct accurate machine-learned potential energy surfaces for both the ground-state electronic surface and excited-state energy gap. We demonstrate the utility of this approach for simulating the dynamics of Nile blue in ethanol, where we experimentally validate and decompose the simulated 2DES to establish the nuclear motions of the chromophore and the solvent that couple to the excited state, connecting the spectroscopic signals to their molecular origin.

physics.chem-ph↗

arXiv2023

SinGRAF: Learning a 3D Generative Radiance Field for a Single Scene

Generative models have shown great promise in synthesizing photorealistic 3D objects, but they require large amounts of training data. We introduce SinGRAF, a 3D-aware generative model that is trained with a few input images of a single scene. Once trained, SinGRAF generates different realizations of this 3D scene that preserve the appearance of the input while varying scene layout. For this purpose, we build on recent progress in 3D GAN architectures and introduce a novel progressive-scale patch discrimination approach during training. With several experiments, we demonstrate that the results produced by SinGRAF outperform the closest related works in both quality and diversity by a large margin.

cs.CV↗

arXiv2023

Temporal Interpolation Is All You Need for Dynamic Neural Radiance Fields

Temporal interpolation often plays a crucial role to learn meaningful representations in dynamic scenes. In this paper, we propose a novel method to train spatiotemporal neural radiance fields of dynamic scenes based on temporal interpolation of feature vectors. Two feature interpolation methods are suggested depending on underlying representations, neural networks or grids. In the neural representation, we extract features from space-time inputs via multiple neural network modules and interpolate them based on time frames. The proposed multi-level feature interpolation network effectively captures features of both short-term and long-term time ranges. In the grid representation, space-time features are learned via four-dimensional hash grids, which remarkably reduces training time. The grid representation shows more than 100 times faster training speed than the previous neural-net-based methods while maintaining the rendering quality. Concatenating static and dynamic features and adding a simple smoothness term further improve the performance of our proposed models. Despite the simplicity of the model architectures, our method achieved state-of-the-art performance both in rendering quality for the neural representation and in training speed for the grid representation.

cs.CV↗

arXiv2020

NPRportrait 1.0: A Three-Level Benchmark for Non-Photorealistic Rendering of Portraits

Despite the recent upsurge of activity in image-based non-photorealistic rendering (NPR), and in particular portrait image stylisation, due to the advent of neural style transfer, the state of performance evaluation in this field is limited, especially compared to the norms in the computer vision and machine learning communities. Unfortunately, the task of evaluating image stylisation is thus far not well defined, since it involves subjective, perceptual and aesthetic aspects. To make progress towards a solution, this paper proposes a new structured, three level, benchmark dataset for the evaluation of stylised portrait images. Rigorous criteria were used for its construction, and its consistency was validated by user studies. Moreover, a new methodology has been developed for evaluating portrait stylisation algorithms, which makes use of the different benchmark levels as well as annotations provided by user studies regarding the characteristics of the faces. We perform evaluation for a wide variety of image stylisation methods (both portrait-specific and general purpose, and also both traditional NPR approaches and neural style transfer) using the new benchmark dataset.

cs.CV↗