cricverse/CricAnnotate · Apply for community grant: Personal project (gpu and storage)

We are a team of three graduate students working on building a foundational model for automatic highlight and commentary generation (both text and audio) for cricket matches worldwide.

So far, we've developed:

A custom dataset tailored for this task
A fine-grained annotation tool for cricket videos
A vision-language model (VLM) capable of generating cricket highlights

To scale our project of multimodal understanding in sports, we require GPU support and additional storage, as we're working with large-scale audio, video, and text data.

We're excited about the opportunity to contribute to the open-source community and would be grateful for the support!