Detection Test Computer Vision Project
Testing
Updated 10 months ago
28
views2
downloadsTags
Metrics
Try This Model
Drop an image or
Description
Experimenting with the platform via ingesting a YouTube video of funny animals. Quickly labeled 100 images with 13 classes (way too few images and too many classes) to see how the platform worked end to end.
Pros
- Upload from YouTube & frame sampling
- Augmentations are an awesome addition.
- Ease of use in an actual app.
- Easier to use than a Jupyter notebook, especially without having to download huge amounts of dependencies.
Cons
- Annotations felt kinda clunky, hard to undo / modify things.
Other
- The model quality wasn't great... But what can you expect from 100 base images and quickly labeling things with waaaay too many classes? Performed as well as I'd expect for low quality data. If I had focused on just one animal I suspect even 100 images would've produced something pretty good.
- 100% delivered on the promise to democratize computer vision. Awesome.
Use This Trained Model
Try it in your browser, or deploy via our Hosted Inference API and other deployment methods.
Build Computer Vision Applications Faster with Supervision
Visualize and process your model results with our reusable computer vision tools.
Cite This Project
LICENSE
CC BY 4.0 If you use this dataset in a research paper, please cite it using the following BibTeX:
@misc{
detection-test-qjcom_dataset,
title = { Detection Test Dataset },
type = { Open Source Dataset },
author = { Testing },
howpublished = { \url{ https://universe.roboflow.com/testing-6bfys/detection-test-qjcom } },
url = { https://universe.roboflow.com/testing-6bfys/detection-test-qjcom },
journal = { Roboflow Universe },
publisher = { Roboflow },
year = { 2024 },
month = { feb },
note = { visited on 2024-11-24 },
}