7 papers
Rank-Aware Hyperbolic Alignment for Vision-Language Dataset Distillation
Jongoh Jeong, Sun-Kyung Lee, Kuk-Jin Yoon
Vision-language dataset distillation (VLDD) compresses a large image-text paired dataset into a small set of synthetic pairs that can efficiently train contrastive vision-language…
Multimodal Distribution Matching for Vision-Language Dataset Distillation
Jongoh Jeong, Hoyong Kwon, Minseok Kim +1
Dataset distillation compresses large training sets into compact synthetic datasets while preserving downstream performance. As modern systems increasingly operate on paired vision…
Improving Black-Box Generative Attacks via Generator Semantic Consistency
Jongoh Jeong, Hunmin Yang, Jaeseok Jeong +1
Transfer attacks optimize on a surrogate and deploy to a black-box target. While iterative optimization attacks in this paradigm are limited by their per-input cost limits efficien…
Task-oriented Learnable Diffusion Timesteps for Universal Few-shot Learning of Dense Tasks
Changgyoon Oh, Jongoh Jeong, Jegyeong Cho +1
Denoising diffusion probabilistic models have brought tremendous advances in generative tasks, achieving state-of-the-art performance thus far. Current diffusion model-based applic…
AVOID: The Adverse Visual Conditions Dataset with Obstacles for Driving Scene Understanding
Jongoh Jeong, Taek-Jin Song, Jong-Hwan Kim +1
Understanding road scenes for visual perception remains crucial for intelligent self-driving cars. In particular, it is desirable to detect unexpected small road hazards reliably i…
Exploring Syn-to-Real Domain Adaptation for Military Target Detection
Jongoh Jeong, Youngjin Oh, Gyeongrae Nam +2
Object detection is one of the key target tasks of interest in the context of civil and military applications. In particular, the real-world deployment of target detection methods…