docs: Update train_model.md to include instructions for training and …

…using custom detection models
locaal-ai · May 13, 2024 · f4f7dab · f4f7dab
1 parent 99f607f
commit f4f7dab
Show file tree

Hide file tree

Showing 3 changed files with 7 additions and 5 deletions.
diff --git a/README.md b/README.md
@@ -52,9 +52,9 @@ Roadmap features:
 - Make available detection information for other plugins through settings
 - More real-time models choices
 
-## Train a custom model
+## Train and use a custom detection model
 
-Follow the instructions in [docs/train_model.md](docs/train_model.md) to train your own custom model.
+Follow the instructions in [docs/train_model.md](docs/train_model.md) to train and use your own custom model.
 
 ## Building
 

diff --git a/docs/image.png b/docs/image.png
diff --git a/docs/train_model.md b/docs/train_model.md
@@ -1,8 +1,10 @@
-# How to train your custom model for OBS Detect plugin
+# How to train and use a custom model with OBS Detect plugin
 
 OBS Detect is based on the [EdgeYOLO](https://github.com/LSH9832/edgeyolo) work.
 They provide a model training script that works with just setting some parameters.
 
+If you already have a trained EdgeYOLO model in `.onnx` format, skip to the [last step](#step-6-use-the-model-with-obs-detect).
+
 You need to get a dataset first. The supported dataset formats are mentiond in the [EdgeYOLO](https://github.com/LSH9832/edgeyolo?tab=readme-ov-file#train) readme: COCO, VOC, YOLO, and DOTA.
 
 In this example we will use a COCO dataset from Roboflow. You can get the dataset from [here](https://public.roboflow.com/object-detection/aquarium/2).
@@ -181,6 +183,6 @@ You will find the ONNX model in the `output/export/` folder, e.g. `output/export
 
 You can now use the ONNX model with the OBS Detect plugin. Just load the model from the plugin settings.
 
-![alt text](image.png)
+![select external model](image.png)
 
-You will also need a configuration file for the model but that is created automatically by the export script above. It will have the same name as the ONNX model but with a `.json` extension.
+You will also need a configuration file for the model with the class names, which is created automatically by the export / conversion script above. It will have the same name as the ONNX model but with a `.json` extension.