halosaber

Title: EcoSplat: Efficiency-controllable Feed-forward 3D Gaussian Splatting from Multi-view Images Authors: Minh-Quan Viet Bui, Jongmin Park, Juan Luis Gonzalez, Jaeho Moon, Jihyong Oh, Munchurl Kim Abstract: Feed-forward 3D Gaussian Splatting (3DGS) enables efficient one-pass scene reconstruction, providing 3D representations for novel view synthesis without per-scene optimization. However, existing methods typically predict pixel-aligned primitives per-view, producing an excessive number of...

Title: OmniText: A Training-Free Generalist for Controllable Text-Image Manipulation Authors: Agus Gunawan, Samuel Teodoro, Yun Chen, Soo Ye Kim, Jihyong Oh, Munchurl Kim Absract: Recent advancements in diffusion-based text synthesis have demonstrated significant performance in inserting and editing text within images via inpainting. However, despite the potential of text inpainting methods, three key limitations hinder their applicability to broader Text Image Manipulation (TIM) tasks:...

Video and Image Computing Lab에서는 2026년 후기 KAIST 국비/KAIST 석사과정생을 모집합니다. Main research areas in VIC Lab are mostly focused but not limited on the following subjects: -3D Image/Video Reconstruction and Recondering * 3D/4D Gaussian Splatting for Statis/Dynamic Scenes     * Neural Radiance Fields (NeRF)    - Image/Video Generative Models     * Multimodal Diffusion Models for Images and Video     * Image/Video Editing and Manipulation using Generative Models     * MMultimodal-based...