Breakdance Video classification in the age of Generative AI
This work addresses a niche but popular domain (breakdance) in sports video analysis, though it is incremental as it applies existing methods to a new dataset.
The study evaluated modern video foundation models for breakdance video classification, finding that video encoder models outperform state-of-the-art video language models in prediction tasks, with specific insights on model selection and decoder analysis.
Large Vision Language models have seen huge application in several sports use-cases recently. Most of these works have been targeted towards a limited subset of popular sports like soccer, cricket, basketball etc; focusing on generative tasks like visual question answering, highlight generation. This work analyzes the applicability of the modern video foundation models (both encoder and decoder) for a very niche but hugely popular dance sports - breakdance. Our results show that Video Encoder models continue to outperform state-of-the-art Video Language Models for prediction tasks. We provide insights on how to choose the encoder model and provide a thorough analysis into the workings of a finetuned decoder model for breakdance video classification.