18 papers
DRIFT: Difficulty-aware Rectified Flows for Through-plane MRI Super-Resolution
Yoonseok Choi, Eun-Gyu Ha, Daniel Kim +3
Magnetic Resonance Imaging (MRI) is often acquired with anisotropic resolution to reduce scan time, producing stair-step artifacts along the through-plane direction. In through-pla…
Controlling Motion Transfer in Diffusion Transformers via Attention Heads
Sunyoung Jung, Jiwoo Park, Yoonseok Choi +3
Diffusion Transformers (DiTs) have advanced video generation with high-quality, temporally coherent results. However, extending them to motion transfer, which requires following re…
SimWorlds: A Multi-Agent System for Dynamic 3D Scene Creation
Chunjiang Liu, Xiaoyuan Wang, Haoyu Chen +3
LLM agents are increasingly used to translate natural language into 3D scenes in a procedural way, but existing systems focus on static output. Dynamic 4D scenes from text alone, i…
Layer-Aware Video Composition via Split-then-Merge
Ozgur Kara, Yujia Chen, Ming-Hsuan Yang +3
We present Split-then-Merge (StM), a novel framework designed to enhance control in generative video composition and address its data scarcity problem. Unlike conventional methods…
Scaling Laws for Deepfake Detection
Wenhao Wang, Longqi Cai, Jusheng Zhang +3
This paper presents a systematic study of scaling laws for the deepfake detection task. Specifically, we analyze the model performance against the number of real image domains, dee…
SceneAdapt: Scene-aware Adaptation of Human Motion Diffusion
Jungbin Cho, Minsu Kim, Jisoo Kim +5
Human motion is inherently diverse and semantically rich, while also shaped by the surrounding scene. However, existing motion generation approaches fail to generate semantically d…
From Prompt to Progression: Taming Video Diffusion Models for Seamless Attribute Transition
Ling Lo, Kelvin C. K. Chan, Wen-Huang Cheng +1
Existing models often struggle with complex temporal changes, particularly when generating videos with gradual attribute transitions. The most common prompt interpolation approach…
Learning Knowledge-based Prompts for Robust 3D Mask Presentation Attack Detection
Fangling Jiang, Qi Li, Bing Liu +4
3D mask presentation attack detection is crucial for protecting face recognition systems against the rising threat of 3D mask attacks. While most existing methods utilize multimoda…
CompleteMe: Reference-based Human Image Completion
Yu-Ju Tsai, Brian Price, Qing Liu +5
Recent methods for human image completion can reconstruct plausible body shapes but often fail to preserve unique details, such as specific clothing patterns or distinctive accesso…
Tuning-Free Image Editing with Fidelity and Editability via Unified Latent Diffusion Model
Qi Mao, Lan Chen, Yuchao Gu +2
Balancing fidelity and editability is essential in text-based image editing (TIE), where failures commonly lead to over- or under-editing issues. Existing methods typically rely on…
CoCoIns: Consistent Subject Generation via Contrastive Instantiated Concepts
Lee Hsin-Ying, Kelvin C. K. Chan, Ming-Hsuan Yang
While text-to-image generative models can synthesize diverse and faithful content, subject variation across multiple generations limits their application to long-form content gener…
HoliSDiP: Image Super-Resolution via Holistic Semantics and Diffusion Prior
Li-Yuan Tsao, Hao-Wei Chen, Hao-Wei Chung +4
Text-to-image diffusion models have emerged as powerful priors for real-world image super-resolution (Real-ISR). However, existing methods may produce unintended results due to noi…
PocoLoco: A Point Cloud Diffusion Model of Human Shape in Loose Clothing
Siddharth Seth, Rishabh Dabral, Diogo Luvizon +4
Modeling a human avatar that can plausibly deform to articulations is an active area of research. We present PocoLoco -- the first template-free, point-based, pose-conditioned gene…
KITTEN: A Knowledge-Intensive Evaluation of Image Generation on Visual Entities
Hsin-Ping Huang, Xinyi Wang, Yonatan Bitton +8
Recent advances in text-to-image generation have improved the quality of synthesized images, but evaluations mainly focus on aesthetics or alignment with text prompts. Thus, it rem…
A Simple Approach to Unifying Diffusion-based Conditional Generation
Xirui Li, Charles Herrmann, Kelvin C. K. Chan +4
Recent progress in image generation has sparked research into controlling these models through condition signals, with various methods addressing specific challenges in conditional…
Re-boosting Self-Collaboration Parallel Prompt GAN for Unsupervised Image Restoration
Xin Lin, Yuyan Zhou, Jingtong Yue +4
Unsupervised restoration approaches based on generative adversarial networks (GANs) offer a promising solution without requiring paired datasets. Yet, these GAN-based approaches st…
Improving Subject-Driven Image Synthesis with Subject-Agnostic Guidance
Kelvin C. K. Chan, Yang Zhao, Xuhui Jia +2
In subject-driven text-to-image synthesis, the synthesis process tends to be heavily influenced by the reference images provided by users, often overlooking crucial attributes deta…
AdaIR: Exploiting Underlying Similarities of Image Restoration Tasks with Adapters
Hao-Wei Chen, Yu-Syuan Xu, Kelvin C. K. Chan +3
Existing image restoration approaches typically employ extensive networks specifically trained for designated degradations. Despite being effective, such methods inevitably entail…