Summary of Storyteller: Improving Long Video Description Through Global Audio-visual Character Identification, by Yichen He et al.
StoryTeller: Improving Long Video Description through Global Audio-Visual Character Identification
by Yichen He, Yuan Lin, Jianchao Wu, Hanchong Zhang, Yuchen Zhang, Ruicheng Le
First submitted to arxiv on: 11 Nov 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a novel system called StoryTeller for generating dense descriptions of long videos, incorporating both visual and audio information. The existing large vision-language models (LVLMs) are limited to processing short videos and struggle with generating coherent descriptions for extended videos. To address this challenge, the authors introduce audio-visual character identification as a key factor, which is essential for consistent video description. StoryTeller uses a multimodal large language model that integrates visual, audio, and text modalities to perform audio-visual character identification on minute-long video clips. The system outperforms all open and closed-source baselines on the MovieStory101 dataset, achieving 9.5% higher accuracy than the strongest baseline, Gemini-1.5-pro. Additionally, incorporating audio-visual character identification from StoryTeller improves the performance of all video description models. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper talks about a new way to describe long videos, like movies, in detail. Currently, computers can’t do this very well because they struggle to understand what’s happening on screen and what people are saying. The authors came up with an idea called StoryTeller that uses a combination of visual, audio, and text information to create accurate descriptions of long videos. They tested their system and found that it did much better than other systems at describing movies in detail. This could be useful for many applications, like helping people who are blind or have difficulty watching videos. |
Keywords
» Artificial intelligence » Gemini » Large language model
