Donghoon Ahn
YOU?
Author Swipe
View article: Vid-CamEdit: Video Camera Trajectory Editing with Generative Rendering from Estimated Geometry
Vid-CamEdit: Video Camera Trajectory Editing with Generative Rendering from Estimated Geometry Open
We introduce Vid-CamEdit, a novel framework for video camera trajectory editing, enabling the re-synthesis of monocular videos along user-defined camera paths. This task is challenging due to its ill-posed nature and the limited multi-view…
View article: A Noise is Worth Diffusion Guidance
A Noise is Worth Diffusion Guidance Open
Diffusion models excel in generating high-quality images. However, current diffusion models struggle to produce reliable images without guidance methods, such as classifier-free guidance (CFG). Are guidance methods truly necessary? Observi…
View article: Dopamine‐conjugated extracellular vesicles induce autophagy in Parkinson's disease
Dopamine‐conjugated extracellular vesicles induce autophagy in Parkinson's disease Open
The application of extracellular vesicles (EVs) as vehicles for anti‐Parkinson's agents represents a significant advance, yet their clinical translation is hampered by challenges in efficient brain delivery and complex blood‐brain barrier …
View article: Geometry-Aware Score Distillation via 3D Consistent Noising and Gradient Consistency Modeling
Geometry-Aware Score Distillation via 3D Consistent Noising and Gradient Consistency Modeling Open
Score distillation sampling (SDS), the methodology in which the score from pretrained 2D diffusion models is distilled into 3D representation, has recently brought significant advancements in text-to-3D generation task. However, this appro…
View article: Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance
Self-Rectifying Diffusion Sampling with Perturbed-Attention Guidance Open
Recent studies have demonstrated that diffusion models are capable of generating high-quality samples, but their quality heavily depends on sampling guidance techniques, such as classifier guidance (CG) and classifier-free guidance (CFG). …
View article: Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generation
Debiasing Scores and Prompts of 2D Diffusion for View-consistent Text-to-3D Generation Open
Existing score-distilling text-to-3D generation techniques, despite their considerable promise, often encounter the view inconsistency problem. One of the most notable issues is the Janus problem, where the most canonical view of an object…
View article: Gαs slow conformational transition upon GTP binding and a novel Gαs regulator
Gαs slow conformational transition upon GTP binding and a novel Gαs regulator Open
SUMMARY G proteins are major signaling partners for G protein-coupled receptors (GPCRs). Although stepwise structural changes during GPCR–G protein complex formation and guanosine diphosphate (GDP) release have been reported, no informatio…
View article: Effect of α-helical domain of Gi/o α subunit on GDP/GTP turnover
Effect of α-helical domain of Gi/o α subunit on GDP/GTP turnover Open
Heterotrimeric guanine nucleotide-binding proteins (G proteins) are composed of α, β, and γ subunits, and Gα has a GDP/GTP-binding pocket. When a guanine nucleotide exchange factor (GEF) interacts with Gα, GDP is released, and GTP interact…
View article: Structural mechanism underlying primary and secondary coupling between GPCRs and the Gi/o family
Structural mechanism underlying primary and secondary coupling between GPCRs and the Gi/o family Open
Heterotrimeric G proteins are categorized into four main families based on their function and sequence, Gs, Gi/o, Gq/11, and G12/13. One receptor can couple to more than one G protein subtype, and the coupling efficiency varies depending o…