administrator/frigate
1
0
Fork 0
mirror of https://github.com/blakeblackshear/frigate.git synced 2026-07-26 21:59:02 +03:00
Code Issues Actions 2 Packages Projects Releases Wiki Activity

refresh the plus documentation (#18462)

Browse Source 
* refresh the plus documentation

* fix broken links
This commit is contained in:
Blake Blackshear
2025-05-30 06:38:25 -05:00
committed by GitHub
parent 948b087d3c
commit ee51326d35
9 changed files with 52 additions and 22 deletions
Download Patch File Download Diff File Expand all files Collapse all files
docs/docs/plus/index.md
+19 -1
View File
@@ -40,9 +40,17 @@ Using Frigate+ models with `onnx` and `rocm` is only available with Frigate 0.15
_\* Requires Frigate 0.15_
## Improving your model
Some users may find that Frigate+ models result in more false positives initially, but by submitting true and false positives, the model will improve. With all the new images now being submitted by subscribers, future base models will improve as more and more examples are incorporated. Note that only images with at least one verified label will be used when training your model. Submitting an image from Frigate as a true or false positive will not verify the image. You still must verify the image in Frigate+ in order for it to be used in training.
- **Submit both true positives and false positives**. This will help the model differentiate between what is and isn't correct. You should aim for a target of 80% true positive submissions and 20% false positives across all of your images. If you are experiencing false positives in a specific area, submitting true positives for any object type near that area in similar lighting conditions will help teach the model what that area looks like when no objects are present.
- **Lower your thresholds a little in order to generate more false/true positives near the threshold value**. For example, if you have some false positives that are scoring at 68% and some true positives scoring at 72%, you can try lowering your threshold to 65% and submitting both true and false positives within that range. This will help the model learn and widen the gap between true and false positive scores.
- **Submit diverse images**. For the best results, you should provide at least 100 verified images per camera. Keep in mind that varying conditions should be included. You will want images from cloudy days, sunny days, dawn, dusk, and night. As circumstances change, you may need to submit new examples to address new types of false positives. For example, the change from summer days to snowy winter days or other changes such as a new grill or patio furniture may require additional examples and training.
## Available label types
Frigate+ models support a more relevant set of objects for security cameras. Currently, the following objects are supported:
Frigate+ models support a more relevant set of objects for security cameras. The labels for annotation in Frigate+ are configurable by editing the camera in the Cameras section of Frigate+. Currently, the following objects are supported:
- **People**: `person`, `face`
- **Vehicles**: `car`, `motorcycle`, `bicycle`, `boat`, `license_plate`
@@ -52,6 +60,16 @@ Frigate+ models support a more relevant set of objects for security cameras. Cur
Other object types available in the default Frigate model are not available. Additional object types will be added in future releases.
### Candidate labels
Candidate labels are also available for annotation. These labels don't have enough data to be included in the model yet, but using them will help add support sooner. You can enable these labels by editing the camera settings.
Where possible, these labels are mapped to existing labels during training. For example, any `baby` labels are mapped to `person` until support for new labels is added.
The candidate labels are: `baby`, `royal mail`, `canada post`, `bpost`, `skunk`, `badger`, `possum`, `rodent`, `kangaroo`, `chicken`, `groundhog`, `boar`, `hedgehog`, `school bus`, `tractor`, `golf cart`, `garbage truck`, `bus`, `sports ball`
Candidate labels are not available for automatic suggestions.
### Label attributes
Frigate has special handling for some labels when using Frigate+ models. `face`, `license_plate`, and delivery logos such as `amazon`, `ups`, and `fedex` are considered attribute labels which are not tracked like regular objects and do not generate review items directly. In addition, the `threshold` filter will have no effect on these labels. You should adjust the `min_score` and other filter values as needed.