Update README for release 1.0.0, add screenshot examples

README.md

@@ -17,7 +17,7 @@ The easiest way to use it is with the [pixi](https://prefix.dev/docs/pixi/overvi
 ### Installation Steps
 
 1. Download the **Source code** (`zip` or `tar.gz`) from the [Releases](https://github.com/humphrem/action/releases) page, or use Git to clone this repo using `git clone https://github.com/humphrem/action.git`
-2. [Install pixi](https://prefix.dev/docs/pixi/overview#installation)
+2. [Install pixi](https://prefix.dev/docs/pixi/overview#installation) using the instructions for your operating system.  NOTE: on Windows, if using the [`iwr` command](https://learn.microsoft.com/en-us/powershell/module/microsoft.powershell.utility/invoke-webrequest?view=powershell-7.3), make sure you are using [PowerShell](https://learn.microsoft.com/en-us/powershell/scripting/overview?view=powershell-7.3) vs. [cmd.exe], or use the [MSI Windows Installer](https://prefix.dev/docs/pixi/overview#windows-installer).
 3. Start a terminal and navigate to the root of the Action project folder you just downloaded or cloned, `cd action`
 4. Enter the command `pixi run setup` to install dependencies and to download the AI models (NOTE: these are large, ~778M, and will take some time to download)
 
@@ -76,7 +76,7 @@ Action can be configured to run in different ways using various arguments and fl
 | `-d`, `--delete-previous-clips` | Whether to delete clips from previous interrupted or old runs before processing a video again. | `--delete-previous-clips` |
 | `-o`, `--output-dir` | Output directory to use for all clips. | `--output-dir ./output` |
 | `-s`, `--show-detections` | Whether to visually show detection frames with bounding boxes. | `--show-detections` |
-| `i`, `--include-bbox-images` | Whether to include the bounding box images for the frames that trigger or extend each detection event, along with the videos in the clips directory. |
+| `-i`, `--include-bbox-images` | Whether to include the bounding box images for the frames that trigger or extend each detection event, along with the videos in the clips directory. | `--include-bbox-images` |
 | `--log-level` | Logging level. Can be `DEBUG`, `INFO`, `WARNING`, or `ERROR`. Defaults to `INFO`. | `--log-level DEBUG` |
 
 > [!NOTE]
@@ -115,3 +115,19 @@ python3 action.py ./video/terrestrial.mov -c 0.45 -m 8.0 -b 2.0 -e terrestrial -
 ```
 
 This would process the file `./video/terrestrial.mov`, use the Megadetector model, set a confidence threshold of `0.45` (i.e. include animal detections with confidence `0.45` and higher), make all clips `8.0` seconds minimum with a `2.0` second buffer added to the start and end of the clip (i.e. `2.0` + `8.0` + `2.0` = `12.0` seconds), and run detections on every 25th frame in the video.
+
+### Example Bounding Box Images
+
+If either of the `-s`/`--show-detections` or `-i`/`--include-bbox-images` flags are included, bounding boxes and confidence scores are also displayed (with `-s`) or written to the clips directory (with `-i`).  These can be helpful when trying to understand what triggered a detection event, or caused it to be extended.
+
+Here are some examples of aquatic and terrestrial bounding box images.
+
+#### Aquatic Examples
+
+![Sample of fish detection](screenshots/fish-example.png)
+![Sample of fish detection](screenshots/fish-example-2.jpg)
+
+#### Terrestrial Examples
+
+![Sample of animal detection](screenshots/animal-example-1.png)
+![Sample of animal detection](screenshots/animal-example-2.jpg)