arXiv subjects

Jintae Park

Publications and source records attributed to Jintae Park.

5 recordsLinked to original sources

arXiv2026

Mechanical Signature of a Chemically Driven Bath

Using optical tweezers, we demonstrate that the CuAAC click reaction increases the positional fluctuations of a trapped 2 micrometer colloid by at least 20% without altering the trap's corner frequency. This excess variance relaxes to the thermal baseline as reagents are depleted. The additional noise is Gaussian, with a flat power spectrum extending past the corner frequency up to 7 kHz. Notably, the noise amplitude decouples from the macroscale reaction rate, and independent molecular force dipoles are orders of magnitude too small to generate the observed variance. Although the chemical free energy released locally exceeds the energy absorbed by the bead by five orders of magnitude, this energy couples inefficiently. The mechanical action of the reacting bath cannot be represented as a sum of independent molecular events.

cond-mat.soft↗

arXiv2026

RECAST: Recasting Vision-Language Semantics into an Actionable Cost Map for Robot Navigation

Safe and robust robot navigation across diverse environments requires a high-level understanding of complex scenes and the ability to carry it into stable motion. Recent works tackle this with learning-based models trained at scale and with approaches built on vision-language models (VLMs). However, learning-based models break down outside their training distribution, while VLM-based approaches bring that understanding but rarely ground it in the scene or align the action with it. To address these limitations, we present RECAST, a robot navigation framework that combines the reasoning of a VLM with the spatial grounding of vision foundation models to build an Actionable Cost map. Given the robot's front view and the user's instruction, we first decompose the scene with the VLM, judging which surfaces are traversable, which objects pose a risk, which heading to prefer, and which gaps are passable. Vision foundation models then ground these surfaces and objects in the image, and all four judgments are spatially recast into one compact cost map. This map both conditions the trajectory decoders and scores their proposals to select the one to execute. As the VLM's answers trail the live scene, both steps draw on cost maps from two points in time: the pivot frame the VLM judged, which carries all four judgments, and the current frame, whose terrain and collision costs are rebuilt from the current image. RECAST improves success over the strongest prior method by 13.3 points in simulation and 31.4 points on a real quadruped, and reduces the collision rate relative to it by 9.6 and 14.3 points, reaching the lowest collision rate among all methods. The project page is available at https://recast-nav.github.io/

cs.RO↗

arXiv2026

ReDesign: Recovering Editable Design Structures from Images via Agentic Decomposition

Recovering an editable design file from a raster image is a common and costly bottleneck in modern design workflows, yet remains challenging since editability depends on recovering multi-modal attributes, such as typography, vector geometry, colors, grouping, and layer ordering. We present ReDesign, an agentic framework that grows an editable layer hierarchy by selecting and composing specialized tools across modalities. To keep this long decision process reliable despite imperfect tool outputs, we introduce graceful verification at each expansion, which provides local accept, prune, or retry feedback that prevents error accumulation and avoids large scale reruns. To evaluate editability at scale, we introduce the Figma Edit Replay Benchmark, consisting of 909 raw Figma files and 14,796 controlled edit instructions that replay edits on reconstructed outputs. Across this benchmark and standard reconstruction metrics, ReDesign achieves strong visual fidelity while delivering the highest editability across layout, color, and text edits, outperforming layered decomposition baselines and serial tool use pipelines.

cs.CV↗

arXiv2025

ODPG: Outfitting Diffusion with Pose Guided Condition

Virtual Try-On (VTON) technology allows users to visualize how clothes would look on them without physically trying them on, gaining traction with the rise of digitalization and online shopping. Traditional VTON methods, often using Generative Adversarial Networks (GANs) and Diffusion models, face challenges in achieving high realism and handling dynamic poses. This paper introduces Outfitting Diffusion with Pose Guided Condition (ODPG), a novel approach that leverages a latent diffusion model with multiple conditioning inputs during the denoising process. By transforming garment, pose, and appearance images into latent features and integrating these features in a UNet-based denoising model, ODPG achieves non-explicit synthesis of garments on dynamically posed human images. Our experiments on the FashionTryOn and a subset of the DeepFashion dataset demonstrate that ODPG generates realistic VTON images with fine-grained texture details across various poses, utilizing an end-to-end architecture without the need for explicit garment warping processes. Future work will focus on generating VTON outputs in video format and on applying our attention mechanism, as detailed in the Method section, to other domains with limited data.

cs.CV↗

arXiv2018

Optical observations of NEA 3200 Phaethon (1983 TB) during the 2017 apparition

The near-Earth asteroid 3200 Phaethon (1983 TB) is an attractive object not only from a scientific viewpoint but also because of JAXA's DESTINY+ target. The rotational lightcurve and spin properties were investigated based on the data obtained in the ground-based observation campaign of Phaethon. We aim to refine the lightcurves and shape model of Phaethon using all available lightcurve datasets obtained via optical observation, as well as our time-series observation data from the 2017 apparition. Using eight 1-2-m telescopes and an optical imager, we acquired the optical lightcurves and derived the spin parameters of Phaethon. We applied the lightcurve inversion method and SAGE algorithm to deduce the convex and non-convex shape model and pole orientations. We analysed the optical lightcurve of Phaethon and derived a synodic and a sidereal rotational period of 3.6039 h, with an axis ratio of a/b = 1.07. The ecliptic longitude (lambda) and latitude (beta) of the pole orientation were determined as (308, -52) and (322, -40) via two independent methods. A non-convex model from the SAGE method, which exhibits a concavity feature, is also presented.

astro-ph.EP↗