arXiv subjects

Younghyun Kim

Publications and source records attributed to Younghyun Kim.

At least 19 recordsLinked to original sources

arXiv2026

3DreamBooth: High-Fidelity 3D Subject-Driven Video Generation Model

Creating dynamic, view-consistent videos of customized subjects is highly sought after for a wide range of emerging applications, including immersive VR/AR, virtual production, and next-generation e-commerce. However, despite rapid progress in subject-driven video generation, existing methods predominantly treat subjects as 2D entities, focusing on transferring identity through single-view visual features or textual prompts. Because real-world subjects are inherently 3D, applying these 2D-centric approaches to 3D object customization reveals a fundamental limitation: they lack the comprehensive spatial priors necessary to reconstruct the 3D geometry. Consequently, when synthesizing novel views, they must rely on generating plausible but arbitrary details for unseen regions, rather than preserving the true 3D identity. Achieving genuine 3D-aware customization remains challenging due to the scarcity of multi-view video datasets. While one might attempt to fine-tune models on limited video sequences, this often leads to temporal overfitting. To resolve these issues, we introduce a novel framework for 3D-aware video customization, comprising 3DreamBooth and 3Dapter. 3DreamBooth decouples spatial geometry from temporal motion through a 1-frame optimization paradigm. By restricting updates to spatial representations, it effectively bakes a robust 3D prior into the model without the need for exhaustive video-based training. To enhance fine-grained textures and accelerate convergence, we incorporate 3Dapter, a visual conditioning module. Following single-view pre-training, 3Dapter undergoes multi-view joint optimization with the main generation branch via an asymmetrical conditioning strategy. This design allows the module to act as a dynamic selective router, querying view-specific geometric hints from a minimal reference set. Project page: https://ko-lani.github.io/3DreamBooth/

cs.CV↗

arXiv2025

Moiré Zero: An Efficient and High-Performance Neural Architecture for Moiré Removal

Moiré patterns, caused by frequency aliasing between fine repetitive structures and a camera sensor's sampling process, have been a significant obstacle in various real-world applications, such as consumer photography and industrial defect inspection. With the advancements in deep learning algorithms, numerous studies-predominantly based on convolutional neural networks-have suggested various solutions to address this issue. Despite these efforts, existing approaches still struggle to effectively eliminate artifacts due to the diverse scales, orientations, and color shifts of moiré patterns, primarily because the constrained receptive field of CNN-based architectures limits their ability to capture the complex characteristics of moiré patterns. In this paper, we propose MZNet, a U-shaped network designed to bring images closer to a 'Moire-Zero' state by effectively removing moiré patterns. It integrates three specialized components: Multi-Scale Dual Attention Block (MSDAB) for extracting and refining multi-scale features, Multi-Shape Large Kernel Convolution Block (MSLKB) for capturing diverse moiré structures, and Feature Fusion-Based Skip Connection for enhancing information flow. Together, these components enhance local texture restoration and large-scale artifact suppression. Experiments on benchmark datasets demonstrate that MZNet achieves state-of-the-art performance on high-resolution datasets and delivers competitive results on lower-resolution dataset, while maintaining a low computational cost, suggesting that it is an efficient and practical solution for real-world applications. Project page: https://sngryonglee.github.io/MoireZero

cs.CV↗

arXiv2025

StarFT: Robust Fine-tuning of Zero-shot Models via Spuriosity Alignment

Learning robust representations from data often requires scale, which has led to the success of recent zero-shot models such as CLIP. However, the obtained robustness can easily be deteriorated when these models are fine-tuned on other downstream tasks (e.g., of smaller scales). Previous works often interpret this phenomenon in the context of domain shift, developing fine-tuning methods that aim to preserve the original domain as much as possible. However, in a different context, fine-tuned models with limited data are also prone to learning features that are spurious to humans, such as background or texture. In this paper, we propose StarFT (Spurious Textual Alignment Regularization), a novel framework for fine-tuning zero-shot models to enhance robustness by preventing them from learning spuriosity. We introduce a regularization that aligns the output distribution for spuriosity-injected labels with the original zero-shot model, ensuring that the model is not induced to extract irrelevant features further from these descriptions. We leverage recent language models to get such spuriosity-injected labels by generating alternative textual descriptions that highlight potentially confounding features. Extensive experiments validate the robust generalization of StarFT and its emerging properties: zero-shot group robustness and improved zero-shot classification. Notably, StarFT boosts both worst-group and average accuracy by 14.30% and 3.02%, respectively, in the Waterbirds group shift scenario, where other robust fine-tuning baselines show even degraded performance.

cs.AI↗

arXiv2025

DiffuseSlide: Training-Free High Frame Rate Video Generation Diffusion

Recent advancements in diffusion models have revolutionized video generation, enabling the creation of high-quality, temporally consistent videos. However, generating high frame-rate (FPS) videos remains a significant challenge due to issues such as flickering and degradation in long sequences, particularly in fast-motion scenarios. Existing methods often suffer from computational inefficiencies and limitations in maintaining video quality over extended frames. In this paper, we present a novel, training-free approach for high FPS video generation using pre-trained diffusion models. Our method, DiffuseSlide, introduces a new pipeline that leverages key frames from low FPS videos and applies innovative techniques, including noise re-injection and sliding window latent denoising, to achieve smooth, consistent video outputs without the need for additional fine-tuning. Through extensive experiments, we demonstrate that our approach significantly improves video quality, offering enhanced temporal coherence and spatial fidelity. The proposed method is not only computationally efficient but also adaptable to various video generation tasks, making it ideal for applications such as virtual reality, video games, and high-quality content creation.

cs.CV↗

arXiv2024

DiffuseHigh: Training-free Progressive High-Resolution Image Synthesis through Structure Guidance

Large-scale generative models, such as text-to-image diffusion models, have garnered widespread attention across diverse domains due to their creative and high-fidelity image generation. Nonetheless, existing large-scale diffusion models are confined to generating images of up to 1K resolution, which is far from meeting the demands of contemporary commercial applications. Directly sampling higher-resolution images often yields results marred by artifacts such as object repetition and distorted shapes. Addressing the aforementioned issues typically necessitates training or fine-tuning models on higher-resolution datasets. However, this poses a formidable challenge due to the difficulty in collecting large-scale high-resolution images and substantial computational resources. While several preceding works have proposed alternatives to bypass the cumbersome training process, they often fail to produce convincing results. In this work, we probe the generative ability of diffusion models at higher resolution beyond their original capability and propose a novel progressive approach that fully utilizes generated low-resolution images to guide the generation of higher-resolution images. Our method obviates the need for additional training or fine-tuning which significantly lowers the burden of computational costs. Extensive experiments and results validate the efficiency and efficacy of our method. Project page: https://yhyun225.github.io/DiffuseHigh/

cs.CV↗

arXiv2024

Discovering and Mitigating Visual Biases through Keyword Explanation

Addressing biases in computer vision models is crucial for real-world AI deployments. However, mitigating visual biases is challenging due to their unexplainable nature, often identified indirectly through visualization or sample statistics, which necessitates additional human supervision for interpretation. To tackle this issue, we propose the Bias-to-Text (B2T) framework, which interprets visual biases as keywords. Specifically, we extract common keywords from the captions of mispredicted images to identify potential biases in the model. We then validate these keywords by measuring their similarity to the mispredicted images using a vision-language scoring model. The keyword explanation form of visual bias offers several advantages, such as a clear group naming for bias discovery and a natural extension for debiasing using these group names. Our experiments demonstrate that B2T can identify known biases, such as gender bias in CelebA, background bias in Waterbirds, and distribution shifts in ImageNet-R/C. Additionally, B2T uncovers novel biases in larger datasets, such as Dollar Street and ImageNet. For example, we discovered a contextual bias between "bee" and "flower" in ImageNet. We also highlight various applications of B2T keywords, including debiased training, CLIP prompting, and model comparison.

cs.LG↗

arXiv2024

I see an IC: A Mixed-Methods Approach to Study Human Problem-Solving Processes in Hardware Reverse Engineering

Trust in digital systems depends on secure hardware, often assured through Hardware Reverse Engineering (HRE). This work develops methods for investigating human problem-solving processes in HRE, an underexplored yet critical aspect. Since reverse engineers rely heavily on visual information, eye tracking holds promise for studying their cognitive processes. To gain further insights, we additionally employ verbal thought protocols during and immediately after HRE tasks: Concurrent and Retrospective Think Aloud. We evaluate the combination of eye tracking and Think Aloud with 41 participants in an HRE simulation. Eye tracking accurately identifies fixations on individual circuit elements and highlights critical components. Based on two use cases, we demonstrate that eye tracking and Think Aloud can complement each other to improve data quality. Our methodological insights can inform future studies in HRE, a specific setting of human-computer interaction, and in other problem-solving settings involving misleading or missing information.

cs.HC↗

arXiv2023

Hydra: Multi-head Low-rank Adaptation for Parameter Efficient Fine-tuning

The recent surge in large-scale foundation models has spurred the development of efficient methods for adapting these models to various downstream tasks. Low-rank adaptation methods, such as LoRA, have gained significant attention due to their outstanding parameter efficiency and no additional inference latency. This paper investigates a more general form of adapter module based on the analysis that parallel and sequential adaptation branches learn novel and general features during fine-tuning, respectively. The proposed method, named Hydra, due to its multi-head computational branches, combines parallel and sequential branch to integrate capabilities, which is more expressive than existing single branch methods and enables the exploration of a broader range of optimal points in the fine-tuning process. In addition, the proposed adaptation method explicitly leverages the pre-trained weights by performing a linear combination of the pre-trained features. It allows the learned features to have better generalization performance across diverse downstream tasks. Furthermore, we perform a comprehensive analysis of the characteristics of each adaptation branch with empirical evidence. Through an extensive range of experiments, encompassing comparisons and ablation studies, we substantiate the efficiency and demonstrate the superior performance of Hydra. This comprehensive evaluation underscores the potential impact and effectiveness of Hydra in a variety of applications. Our code is available on \url{https://github.com/extremebird/Hydra}

cs.CV↗

arXiv2023

ORA3D: Overlap Region Aware Multi-view 3D Object Detection

Current multi-view 3D object detection methods often fail to detect objects in the overlap region properly, and the networks' understanding of the scene is often limited to that of a monocular detection network. Moreover, objects in the overlap region are often largely occluded or suffer from deformation due to camera distortion, causing a domain shift. To mitigate this issue, we propose using the following two main modules: (1) Stereo Disparity Estimation for Weak Depth Supervision and (2) Adversarial Overlap Region Discriminator. The former utilizes the traditional stereo disparity estimation method to obtain reliable disparity information from the overlap region. Given the disparity estimates as supervision, we propose regularizing the network to fully utilize the geometric potential of binocular images and improve the overall detection accuracy accordingly. Further, the latter module minimizes the representational gap between non-overlap and overlapping regions. We demonstrate the effectiveness of the proposed method with the nuScenes large-scale multi-view 3D object detection data. Our experiments show that our proposed method outperforms current state-of-the-art models, i.e., DETR3D and BEVDet.

cs.CV↗

arXiv2023

How Interactions Influence Users' Security Perception of Virtual Reality Authentication?

Users readily embrace the rapid advancements in virtual reality (VR) technology within various everyday contexts, such as gaming, social interactions, shopping, and commerce. In order to facilitate transactions and payments, VR systems require access to sensitive user data and assets, which consequently necessitates user authentication. However, there exists a limited understanding regarding how users' unique experiences in VR contribute to their perception of security. In our study, we adopt a research approach known as ``technology probe'' to investigate this question. Specifically, we have designed probes that explore the authentication process in VR, aiming to elicit responses from participants from multiple perspectives. These probes were seamlessly integrated into the routine payment system of a VR game, thereby establishing an organic study environment. Through qualitative analysis, we uncover the interplay between participants' interaction experiences and their security perception. Remarkably, despite encountering unique challenges in usability during VR interactions, our participants found the intuitive virtualized authentication process beneficial and thoroughly enjoyed the immersive nature of VR. Furthermore, we observe how these interaction experiences influence participants' ability to transfer their pre-existing understanding of authentication into VR, resulting in a discrepancy in perceived security. Moreover, we identify users' conflicting expectations, encompassing their desire for an enjoyable VR experience alongside the assurance of secure VR authentication. Building upon our findings, we propose recommendations aimed at addressing these expectations and alleviating potential conflicts.

cs.CR↗

arXiv2021

Moonshine: An Online Randomness Distiller for Zero-Involvement Authentication

Context-based authentication is a method for transparently validating another device's legitimacy to join a network based on location. Devices can pair with one another by continuously harvesting environmental noise to generate a random key with no user involvement. However, there are gaps in our understanding of the theoretical limitations of environmental noise harvesting, making it difficult for researchers to build efficient algorithms for sampling environmental noise and distilling keys from that noise. This work explores the information-theoretic capacity of context-based authentication mechanisms to generate random bit strings from environmental noise sources with known properties. Using only mild assumptions about the source process's characteristics, we demonstrate that commonly-used bit extraction algorithms extract only about 10% of the available randomness from a source noise process. We present an efficient algorithm to improve the quality of keys generated by context-based methods and evaluate it on real key extraction hardware. Moonshine is a randomness distiller which is more efficient at extracting bits from an environmental entropy source than existing methods. Our techniques nearly double the quality of keys as measured by the NIST test suite, producing keys that can be used in real-world authentication scenarios.

cs.CR↗

arXiv2021

HW/SW Framework for Improving the Safety of Implantable and Wearable Medical Devices

Implantable and wearable medical devices (IWMDs) are widely used for the monitoring and therapy of an increasing range of medical conditions. Improvements in medical devices, enabled by advances in low-power processors, more complex firmware, and wireless connectivity, have greatly improved therapeutic outcomes and patients' quality-of-life. However, security attacks, malfunctions and sometimes user errors have raised great concerns regarding the safety of IWMDs. In this work, we present a HW/SW (Hardware/Software) framework for improving the safety of IWMDs, wherein a set of safety rules and a rule check mechanism are used to monitor both the extrinsic state (the patient's physiological parameters sensed by the IWMD) and the internal state of the IWMD (I/O activities of the microcontroller) to infer unsafe operations that may be triggered by user errors, software bugs, or security attacks. We discuss how this approach can be realized in the context of a artificial pancreas with wireless connectivity and implement a prototype to demonstrate its effectiveness in improving safety at modest overheads.

cs.CR↗

arXiv2017

Transport studies of epi-Al/InAs 2DEG systems for required building-blocks in topological superconductor networks

One-dimensional (1D) electronic transport and induced superconductivity in semiconductor nano-structures are crucial ingredients to realize topological superconductivity. Our approach for topological superconductivity employs a two-dimensional electron gas (2DEG) formed by an InAs quantum well, cleanly interfaced with a superconductor (epitaxial Al). This epi-Al/InAs quantum well heterostructure is advantageous for fabricating large-scale nano-structures consisting of multiple Majorana zero modes. Here, we demonstrate building-block transport studies using a high-quality epi-Al/InAs 2DEG heterostructure, which could be put together to realize the proposed 1D nanowire-based nano-structures and 2DEG-based networks that could host multiple Majorana zero modes: 1D transport using 1) quantum point contacts and 2) gate-defined quasi-1D channels in the InAs 2DEG as well as induced superconductivity in 3) a ballistic Al-InAs 2DEG-Al Josephson junction. From 1D transport, systematic evolution of conductance plateaus in half-integer conductance quanta are observed as a result of strong spin-orbit coupling in the InAs 2DEG. Large IcRn, a product of critical current and normal state resistance from the Josephson junction, indicates that the interface between the epitaxial Al and the InAs 2DEG is highly transparent. Our results of electronic transport studies based on the 2D approach suggest that the epitaxial superconductor/2D semiconductor system is suitable for realizing large-scale nano-structures for quantum computing applications.

cond-mat.mes-hall↗

arXiv2017

Coulomb Blockade in Fractional Topological Superconductors

We study charge transport through a floating mesoscopic superconductor coupled to counterpropagating fractional quantum Hall edges at filling fraction $ν=2/3$. We consider a superconducting island with finite charging energy and investigate its effect on transport through the device. We calculate conductance through such a system as a function of temperature and gate voltage applied to the superconducting island. We show that transport is strongly affected by the presence of parafermionic zero modes, leading at zero temperature to a zero-bias conductance quantized in units of $νe^2/h$ independent of the applied gate voltage.

cond-mat.str-el↗

arXiv2016

Signatures of Majorana Kramers pairs in superconductor-Luttinger liquid and superconductor-quantum dot-normal lead junctions

Time-reversal invariant topological superconductors are characterized by the presence of Majorana Kramers pairs localized at defects. One of the transport signatures of Majorana Kramers pairs is the quantized differential conductance of $4e^2/h$ when such a one-dimensional superconductor is coupled to a normal-metal lead. The resonant Andreev reflection, responsible for this phenomenon, can be understood as the boundary condition change for lead electrons at low energies. In this paper, we study the stability of the Andreev reflection fixed point with respect to electron-electron interactions in the Luttinger liquid. We first calculate the phase diagram for the Luttinger liquid-Majorana Kramers pair junction and show that its low-energy properties are determined by Andreev reflection scattering processes in the spin-triplet channel, i.e. the corresponding Andreev boundary conditions are similar to that in a spin-triplet superconductor - normal lead junction. We also study here a quantum dot coupled to a normal lead and a Majorana Kramers pair and investigate the effect of local repulsive interactions leading to an interplay between Kondo and Majorana correlations. Using a combination of renormalization group analysis and slave-boson mean-field theory, we show that the system flows to a new fixed point which is controlled by the Majorana interaction rather than the Kondo coupling. This Majorana fixed point is characterized by correlations between the localized spin and the fermion parity of each spin sector of the topological superconductor. We investigate the stability of the Majorana phase with respect to Gaussian fluctuations.

cond-mat.mes-hall↗

arXiv2016

Gate Tunable In- and Out-Of-Plane Spin-Orbit Coupling and Spin Splitting Anisotropy at LaAlO3/SrTiO3 (110) Interface

Manipulating spin-orbit coupling (SOC) is important for devices such as spin-orbit torque based memory and its understanding is neccessary to answer several fundamental open questions in triplet state superconductivity, topological insulators and Majorana fermions. Here we report spin splitting of 25 meV at the LaAlO3/SrTiO3 (110) interface for in-plane spins at a current density of 1.4x104 A/cm2, which is large compared to that found in semiconductor heterostructures or the LaAlO3/SrTiO3 (100) interface, and in addition it is anisotropic. The anisotropy arises from the difference in electron effective mass along the [001] and [1-10] directions. Our study predicts a spin splitting energy > 1000 meV at a current density of 107 A/cm2, which is enormous compared to metallic systems and will be an ideal spin polarized source. In addition to the in-plane effect, there is an unexpected gate-tunable out-of-plane SOC at the LaAlO3/SrTiO3 (110) interface when the spins lie out-of-plane due to broken symmetry in the plane of the interface. We demonstrate that this can be manipulated by varying the LaAlO3 thickness showing that this interface can be engineered for spin-orbit torque devices.

cond-mat.mes-hall↗

arXiv2015

Topological superconductivity in a multichannel Yu-Shiba-Rusinov chain

Chains of magnetic atoms placed on the surface of an s-wave superconductor with large spin-orbit coupling provide a promising platform for the realization of topological superconducting states characterized by the presence of Majorana zero-energy modes. In this work we study the properties of the one-dimensional chain of Yu-Shiba-Rusinov states induced by magnetic impurities using a realistic model for the magnetic atoms that include the presence of multiple scattering channels. These channels are mixed by the spin-orbit coupling and, via the hybridization of the Yu-Shiba-Rusinov states at different sites of the chain, result in a multi-band structure for the chain. We obtain the topological phase diagram for such band structure. We identify the parameter regimes for which the different bands lead to a topological phase and show that the inclusion of higher bands can greatly enlarge the phase space for the realization of topological states.

cond-mat.supr-con↗

arXiv2015

Two-dimensional epitaxial superconductor-semiconductor heterostructures: A platform for topological superconducting networks

Progress in the emergent field of topological superconductivity relies on synthesis of new material combinations, combining superconductivity, low density, and spin-orbit coupling (SOC). For example, theory [1-4] indicates that the interface between a one-dimensional (1D) semiconductor (Sm) with strong SOC and a superconductor (S) hosts Majorana modes with nontrivial topological properties [5-8]. Recently, epitaxial growth of Al on InAs nanowires was shown to yield a high quality S-Sm system with uniformly transparent interfaces [9] and a hard induced gap, indicted by strongly suppressed sub gap tunneling conductance [10]. Here we report the realization of a two-dimensional (2D) InAs/InGaAs heterostructure with epitaxial Al, yielding a planar S-Sm system with structural and transport characteristics as good as the epitaxial wires. The realization of 2D epitaxial S-Sm systems represent a significant advance over wires, allowing extended networks via top-down processing. Among numerous potential applications, this new material system can serve as a platform for complex networks of topological superconductors with gate-controlled Majorana zero modes [1-4]. We demonstrate gateable Josephson junctions and a highly transparent 2D S-Sm interface based on the product of excess current and normal state resistance.

cond-mat.mes-hall↗