Pretrained Object Detection Models (YOLOv5 & YOLOv8)

Dataset Colab now has pretrained object detection models! Download the model weights and follow our documentation to start detecting notes and robots. We currently have 4 pretrained models:

  • YOLOv5n (95.92% mAP)
  • YOLOv5s (96.01% mAP)
  • YOLOv8n (96.25% mAP)
  • YOLOv8s (95.60% mAP)

On 2/1/2024, we will be releasing Limelight + Coral compatible (tensorflow) models: ssd-mobilenet-v2, efficientdet-d0, and ssd-mobilenet-v2-fpnlite-320. Signup to get an email notification when these are released.

9 Likes

Update: Dataset Colab now has only note and only robot models in addition to the note and robot models. Visit the models page to download and run these new models.

New Google Coral Compatible Tensorflow Models

Dataset Colab now has Tensorflow object detection models (SSD Mobilenet v2 & EfficientDet) that can be used with the Google Coral and a variety of other accelerators! To download and get started using these models visit our models page to download the model and our documentation page for a guide to get started.

  • SSD Mobilenet v2 (92.20% mAP50)
  • EfficientDet (94.50% mAP50)

New Only-Note and Only-Robot Models

In addition to these models we also trained only-note and only-robot models. These models are more specialized and have greater performance. To download and use these models visit our models page .

Recommend New Models

If there are any other models that you would like to see on Dataset Colab, DM me and we would be happy to train them for you!

2 Likes

It would be nice to differentiate between red and blue robots.

Yes! We will do this next year. Thank you!

3 Likes

So I haven’t tried these models. I am curious do you need to see the ‘hole’ for note detection or will an orange rounded rectangle also detect (ie a note from a low angle)

This year we trained our own model because the Limelight model needed the hole and our angle was low so all the notes looked like orange rounded rectangles.

Our models do not need to see the hole in the note to detect it. If you have a camera at a low angle it will still detect the note.

1 Like

Any chance of having that in the offseason?

Hello, there seems to be a problem with your link.(models page)

Sorry about that. When we converted Dataset Colab (only FRC) to Synodic (non-FRC), we had to reconfigure the links. We are still working to reupload the FRC datasets to Synodic. I can keep you updated and provide new links when everything is back to normal.

4 Likes

Ahhhh, i just went to redownload the sat wsbd was wodnerig what happened to it since everythig just linked back to ancempy synodic lmao