Do We Need Large VLMs for Spotting Soccer Actions?
This work addresses the need for lightweight and scalable action spotting in soccer, offering a novel approach that reduces computational costs, though it is incremental in its domain-specific application.
The paper tackles the problem of soccer action spotting by shifting from video-centric to text-based methods using Large Language Models (LLMs) instead of Vision-Language Models (VLMs), achieving performance close to state-of-the-art video-based spotters with zero video processing compute and similar time requirements.
Traditional video-based tasks like soccer action spotting rely heavily on visual inputs, often requiring complex and computationally expensive models to process dense video data. We propose a shift from this video-centric approach to a text-based task, making it lightweight and scalable by utilizing Large Language Models (LLMs) instead of Vision-Language Models (VLMs). We posit that expert commentary, which provides rich descriptions and contextual cues contains sufficient information to reliably spot key actions in a match. To demonstrate this, we employ a system of three LLMs acting as judges specializing in outcome, excitement, and tactics for spotting actions in soccer matches. Our experiments show that this language-centric approach performs effectively in detecting critical match events coming close to state-of-the-art video-based spotters while using zero video processing compute and similar amount of time to process the entire match.