Title: Looking for Help: Building a Cricket Shot Detection Model with YOLOv (Using IPL Footage)

Hey devs, I’m working on a personal project to detect cricket shots from broadcast footage — and I’m looking for someone to collaborate with, especially around training the dataset. Project Goal: Use YOLOv (You Only Look Once) to detect different types of cricket shots (like cover drive, pull, sweep, etc.) from IPL broadcast videos. The idea is to train a model that can recognize these shots based on pose, bat movement, and other visual cues. Where I’m At: I’ve chosen YOLOv as the object detection framework. Planning to extract and label frames from IPL footage. Need to build a custom dataset with accurate annotations. What I Need Help With: Annotating and prepping the dataset (frame extraction, bounding boxes, labeling). Training the YOLOv model. (Optional) Model evaluation and fine-tuning. If you have experience with: YOLOv (any version) Computer vision / deep learning Video annotation tools (like CVAT or Roboflow) Or just a passion for cricket + AI I’d love to team up and build this out together. Next Steps: Drop a comment or DM me if you're interested! We can sync up and figure out the best way to split the work. Happy to share details, plan, and existing materials. Let’s make something cool for cricket fans and ML geeks! Cheers, Krithick

Apr 20, 2025 - 07:13

Title: Looking for Help: Building a Cricket Shot Detection Model with YOLOv (Using IPL Footage)

Hey devs,

I’m working on a personal project to detect cricket shots from broadcast footage — and I’m looking for someone to collaborate with, especially around training the dataset.

Project Goal: Use YOLOv (You Only Look Once) to detect different types of cricket shots (like cover drive, pull, sweep, etc.) from IPL broadcast videos. The idea is to train a model that can recognize these shots based on pose, bat movement, and other visual cues.

Where I’m At:

I’ve chosen YOLOv as the object detection framework.

Planning to extract and label frames from IPL footage.

Need to build a custom dataset with accurate annotations.

What I Need Help With:

Annotating and prepping the dataset (frame extraction, bounding boxes, labeling).

Training the YOLOv model.

(Optional) Model evaluation and fine-tuning.

If you have experience with:

YOLOv (any version)

Computer vision / deep learning

Video annotation tools (like CVAT or Roboflow)

Or just a passion for cricket + AI

I’d love to team up and build this out together.

Next Steps: Drop a comment or DM me if you're interested! We can sync up and figure out the best way to split the work. Happy to share details, plan, and existing materials.

Let’s make something cool for cricket fans and ML geeks!

Cheers,
Krithick