Number it: Temporal Grounding Videos like Flipping Manga
This addresses a specific bottleneck in video understanding for researchers and practitioners in computer vision, offering an incremental but effective enhancement to existing Vid-LLMs.
The paper tackles the problem of Video Temporal Grounding (VTG), where Video Large Language Models (Vid-LLMs) struggle with precise temporal localization, by introducing Number-Prompt (NumPro), a method that adds numerical identifiers to video frames to improve performance. The result shows significant boosts, with up to 6.9% improvement in mIoU for moment retrieval and 8.5% in mAP for highlight detection, setting a new state-of-the-art.
Video Large Language Models (Vid-LLMs) have made remarkable advancements in comprehending video content for QA dialogue. However, they struggle to extend this visual understanding to tasks requiring precise temporal localization, known as Video Temporal Grounding (VTG). To address this gap, we introduce Number-Prompt (NumPro), a novel method that empowers Vid-LLMs to bridge visual comprehension with temporal grounding by adding unique numerical identifiers to each video frame. Treating a video as a sequence of numbered frame images, NumPro transforms VTG into an intuitive process: flipping through manga panels in sequence. This allows Vid-LLMs to "read" event timelines, accurately linking visual content with corresponding temporal information. Our experiments demonstrate that NumPro significantly boosts VTG performance of top-tier Vid-LLMs without additional computational cost. Furthermore, fine-tuning on a NumPro-enhanced dataset defines a new state-of-the-art for VTG, surpassing previous top-performing methods by up to 6.9\% in mIoU for moment retrieval and 8.5\% in mAP for highlight detection. The code will be available at https://github.com/yongliang-wu/NumPro.