6 papers
Grounding Confusable Movie Moments
Dong-Hee Kim, Seonwoo Choi, Changbeen Kim +5
Characters, locations, and actions recur throughout a movie, so many of its moments differ only in fine-grained details. Grounding such a moment in a full-length movie requires res…
Beyond Visual Boundaries: Rethinking Scene Segmentation for Movie RAG
Dong-Hee Kim, Seonwoo Choi, Changbeen Kim +6
Understanding long-form video remains a fundamental challenge for multimodal large language models (MLLMs). Sparse frame sampling fails to capture fine-grained visual details, whil…
Correlation-Weighted Multi-Reward Optimization for Compositional Generation
Jungmyung Wi, Hyunsoo Kim, Donghyun Kim
Text-to-image models produce images that align well with natural language prompts, but compositional generation has long been a central challenge. Models often struggle to satisfy…
CREPE: Coordinate-Aware End-to-End Document Parser
Yamato Okamoto, Youngmin Baek, Geewook Kim +5
In this study, we formulate an OCR-free sequence generation model for visual document understanding (VDU). Our model not only parses text from document images but also extracts the…
SCOB: Universal Text Understanding via Character-wise Supervised Contrastive Learning with Online Text Rendering for Bridging Domain Gap
Daehee Kim, Yoonsik Kim, DongHyun Kim +3
Inspired by the great success of language model (LM)-based pre-training, recent studies in visual document understanding have explored LM-based pre-training methods for modeling te…
On Web-based Visual Corpus Construction for Visual Document Understanding
Donghyun Kim, Teakgyu Hong, Moonbin Yim +2
In recent years, research on visual document understanding (VDU) has grown significantly, with a particular emphasis on the development of self-supervised learning methods. However…