Add show_anns function

Author: giswqs

docs/examples/fast_sam.ipynb

@@ -10,6 +10,8 @@
     "[![image](https://img.shields.io/badge/Open-Planetary%20Computer-black?style=flat&logo=microsoft)](https://pccompute.westeurope.cloudapp.azure.com/compute/hub/user-redirect/git-pull?repo=https://github.com/opengeos/segment-geospatial&urlpath=lab/tree/segment-geospatial/docs/examples/fast_sam.ipynb&branch=main)\n",
     "[![image](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/opengeos/segment-geospatial/blob/main/docs/examples/fast_sam.ipynb)\n",
     "\n",
+    "FastSAM: https://github.com/CASIA-IVA-Lab/FastSAM\n",
+    "\n",
     "Make sure you use GPU runtime for this notebook. For Google Colab, go to `Runtime` -> `Change runtime type` and select `GPU` as the hardware accelerator. "
    ]
   },
@@ -143,6 +145,13 @@
     "sam = SamGeo(model=\"FastSAM-x.pt\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Set the image."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -152,6 +161,13 @@
     "sam.set_image(\"Image.tif\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Segment the image with `everything_prompt`. You can also try `point_prompt`, `box_prompt`, or `text_prompt`."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -161,6 +177,36 @@
     "sam.everything_prompt(output=\"mask.tif\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Show the annotated image."
+   ]
+  },
+  {
+   "cell_type": "code",
+   "execution_count": null,
+   "metadata": {},
+   "outputs": [],
+   "source": [
+    "sam.show_anns(\"mask.png\")"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "![](https://i.imgur.com/af4bj7O.png)"
+   ]
+  },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Convert the segmentation results from GeoTIFF to vector."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,
@@ -170,6 +216,13 @@
     "sam.raster_to_vector(\"mask.tif\", \"mask.geojson\")"
    ]
   },
+  {
+   "cell_type": "markdown",
+   "metadata": {},
+   "source": [
+    "Show the segmentation results on the map."
+   ]
+  },
   {
    "cell_type": "code",
    "execution_count": null,

docs/installation.md

@@ -23,7 +23,7 @@ mamba install -c conda-forge segment-geospatial
 Samgeo-geospatial has some optional dependencies that are not included in the default conda environment. To install these dependencies, run the following command:
 
 ```bash
-mamba install -c conda-forge groundingdino-py
+mamba install -c conda-forge groundingdino-py segment-anything-fast
 ```
 
 As of July 9th, 2023 Linux systems have also required that `libgl1` be installed for segment-geospatial to work. The following command will install that dependency

requirements_dev.txt

@@ -1,2 +1,2 @@
 rio-cogeo
-segment-anything-hq
+segment-anything-fast

samgeo/fast_sam.py

@@ -285,3 +285,28 @@ def raster_to_vector(
             dst_crs=dst_crs,
             **kwargs,
         )
+
+    def show_anns(
+        self,
+        output=None,
+        **kwargs,
+    ):
+        """Show the annotations (objects with random color) on the input image.
+
+        Args:
+            figsize (tuple, optional): The figure size. Defaults to (12, 10).
+            axis (str, optional): Whether to show the axis. Defaults to "off".
+            alpha (float, optional): The alpha value for the annotations. Defaults to 0.35.
+            output (str, optional): The path to the output image. Defaults to None.
+            blend (bool, optional): Whether to show the input image. Defaults to True.
+        """
+
+        annotations = self.annotations
+        prompt_process = self.prompt_process
+
+        if output is None:
+            output = temp_file_path(".png")
+
+        prompt_process.plot(annotations, output, **kwargs)
+
+        show_image(output)