9 papers · 1 filter
Improving Sparse-View 3DGS Generalization via Flat Minima Optimization
Kangmin Seo, Sangeek Hyun, MinKyu Lee +1
Recent advances in neural rendering have established 3D Gaussian Splatting (3DGS) as a highly efficient representation for novel view synthesis, enabling fast training and real-tim…
Cross-scale Aligned Supervision for Training GANs
Sangeek Hyun, MinKyu Lee, Jae-Pil Heo
Modern GANs often introduce adversarial supervision on intermediate generator outputs and interpret the resulting multi-stage synthesis as coarse-to-fine hierarchical generation. I…
PDF-GS: Progressive Distractor Filtering for Robust 3D Gaussian Splatting
Kangmin Seo, MinKyu Lee, Tae-Young Kim +3
Recent advances in 3D Gaussian Splatting (3DGS) have enabled impressive real-time photorealistic rendering. However, conventional training pipelines inherently assume full multi-vi…
SeaCache: Spectral-Evolution-Aware Cache for Accelerating Diffusion Models
Jiwoo Chung, Sangeek Hyun, MinKyu Lee +5
Diffusion models are a strong backbone for visual generation, but their inherently sequential denoising process leads to slow inference. Previous methods accelerate sampling by cac…
Scalable GANs with Transformers
Sangeek Hyun, MinKyu Lee, Jae-Pil Heo
Scalability has driven recent advances in generative modeling, yet its principles remain underexplored for adversarial learning. We investigate the scalability of Generative Advers…
Analyzing the Training Dynamics of Image Restoration Transformers: A Revisit to Layer Normalization
MinKyu Lee, Sangeek Hyun, Woojin Jun +3
This work analyzes the training dynamics of Image Restoration (IR) Transformers and uncovers a critical yet overlooked issue: conventional LayerNorm (LN) drives feature magnitudes…
Fine-Tuning Visual Autoregressive Models for Subject-Driven Generation
Jiwoo Chung, Sangeek Hyun, Hyunjun Kim +3
Recent advances in text-to-image generative models have enabled numerous practical applications, including subject-driven generation, which fine-tunes pretrained models to capture…
Auto-Encoded Supervision for Perceptual Image Super-Resolution
MinKyu Lee, Sangeek Hyun, Woojin Jun +1
This work tackles the fidelity objective in the perceptual super-resolution~(SR). Specifically, we address the shortcomings of pixel-level loss ($\mathcal{L}_\text{pix…
Noise-free Optimization in Early Training Steps for Image Super-Resolution
MinKyu Lee, Jae-Pil Heo
Recent deep-learning-based single image super-resolution (SISR) methods have shown impressive performance whereas typical methods train their networks by minimizing the pixel-wise…