4 papers
DLIOS: An LLM-Augmented Real-Time Multi-Modal Interactive Enhancement Overlay System for Douyin Live Streaming
Shuide Wen, Sungil Seok, Beier Ku +6
We present DLIOS, a Large Language Model (LLM)-augmented real-time multi-modal interactive enhancement overlay system for Douyin (TikTok) live streaming. DLIOS employs a three-laye…
IE-Critic-R1: Advancing the Explanatory Measurement of Text-Driven Image Editing for Human Perception Alignment
Bowen Qu, Shangkun Sun, Xiaoyu Liang +1
Recent advances in text-driven image editing have been significant, yet the task of accurately evaluating these edited images continues to pose a considerable challenge. Different…
Content-Rich AIGC Video Quality Assessment via Intricate Text Alignment and Motion-Aware Consistency
Shangkun Sun, Xiaoyu Liang, Bowen Qu +1
The advent of next-generation video generation models like \textit{Sora} poses challenges for AI-generated content (AIGC) video quality assessment (VQA). These models substantially…
IE-Bench: Advancing the Measurement of Text-Driven Image Editing for Human Perception Alignment
Shangkun Sun, Bowen Qu, Xiaoyu Liang +2
Recent advances in text-driven image editing have been significant, yet the task of accurately evaluating these edited images continues to pose a considerable challenge. Different…