Moved capture_sequence.py to marimapper.py

removed reconstructor entirely
fixed remesher, this needs to be moved into marimapper
fixed view_2d_map

README.md

@@ -10,10 +10,6 @@
 
 This is a selection of tools to map LEDs into 2D and 3D space using only your webcam!
 
-> [!CAUTION]
-> This tool is currently undergoing a rework (10/06/2024) and therefore some features such as rendering strips and cameras may not be available
-
-
 > [!TIP]
 > All scripts can be run with the `--help` argument to list optional parameters such as resolution and exposure.
 
@@ -75,8 +71,8 @@ class Backend:
         pass
 
 ```
-
-You can test your backend with `python scripts/check_backend.py`
+> [!TIP]
+> You can test your backend with `python scripts/check_backend.py`
 
 MariMapper also support the following pre-made backends. This can be selected in the following steps using the `--backend`
 argument.
@@ -87,32 +83,29 @@ argument.
 - [ ] [LCM](https://lcm-proj.github.io/lcm/)
 
 
-## Step 3: Capture a 2D map
+## Step 3: Start scanning!
 
-Set up your LEDs in front of your camera and
-run `python scripts/capture_sequence.py my_scan --backend fadecandy`
-
-Change `--backend` to whatever backend you're using
+[It's time to thunderize!](https://youtu.be/-5KJiHc3Nuc?t=121)
 
-This will produce a timestamped CSV file in the `my_scan` folder with led index, u and v values.
+run `python marimapper.py my_scan --backend fadecandy`
+Change `fadecandy` to whatever backend you're using and `my_scan` to whatever you want to call your scan
 
-Run `python scripts/visualise.py <filename>` to visualise 2D or 3D map files.
+Set up your LEDs so most of them are in view and when you're ready, type `y` when prompted with `Start scan? [y/n]`
 
-## Step 4: Construct a 3D map
+This will turn each LED on and off in turn, do not move the camera or leds during capture!
 
-[It's time to thunderize!](https://youtu.be/-5KJiHc3Nuc?t=121)
+If you just want a 2D map, this is where you can stop! 
+Run `python scripts/view_2d_map.py my_scan/...` to visualise your map replacing `...` with the map name.
 
-To create a 3D map, run `capture_sequence` multiple times from different views of your LEDs,
-this can either be done by moving your webcam around your LEDs or rotating your LEDs.
+To capture a 3D map, rotate your leds or move your webcam to a new position
 
 > [!TIP]
-> You can move your webcam to wherever you like as long as some of your leds are mostly in view
+> As long as some of your leds are mostly in view, you can move your webcam to wherever you like!
 > Try and get at least 3 views between 6° - 20° apart
 
-Once you have a selection of 2D maps captured with the `capture_sequence.py` script,
-run `python scripts/reconstruct.py my_scan`
+Once you have at least 2 2d maps, a new window will appear showing the reconstructed 3D positions of your LEDs.
 
-This may take a while, however once complete will generate `reconstruction.csv` in the `my_scan` folder.
+If it doesn't look quite right, add some more scans!
 
 Here is an example reconstruction of Highbeam's body LEDs
 
@@ -129,17 +122,6 @@ Here is an example reconstruction of Highbeam's body LEDs
 - Use `1`, `2` & `3` keys to change colour scheme
 </details>
 
-
-## Step 5: Construct a mesh (optional)
-
-If you have a high enough density 3d map, you can use the remesh tool to create a 3D mesh based on your leds!
-
-Run `python scripts/remesh.py reconstruction.csv my_mesh.ply`
-
-This will generate a ply file which you can open and look at with your eyes
-
-![alt text](docs/images/remesh_with_normals.png "Highbeam LED mesh reconstruction")
-
 # Feedback
 
 I would really love to hear what you think and if you have any bugs or improvements, please raise them here or drop me a

lib/remesher.py

@@ -41,12 +41,16 @@ def remesh(led_map, mesh_detail=8):
 
     pcd = open3d.geometry.PointCloud()
 
-    pcd.points = open3d.utility.Vector3dVector(
-        [led_map[led_id]["pos"] for led_id in led_map.data]
-    )
-    pcd.normals = open3d.utility.Vector3dVector(
-        [led_map[led_id]["normal"] for led_id in led_map.data]
-    )
+    xyz = []
+    normals = []
+    for led_id in led_map.keys():
+        xyz.append(led_map[led_id]["pos"])
+        normals.append(
+            led_map[led_id]["normal"] / np.linalg.norm(led_map[led_id]["normal"])
+        )
+
+    pcd.points = open3d.utility.Vector3dVector(xyz)
+    pcd.normals = open3d.utility.Vector3dVector(normals)
 
     with open3d.utility.VerbosityContextManager(
         open3d.utility.VerbosityLevel.Debug

scripts/capture_sequence.py → marimapper.py

@@ -1,12 +1,9 @@
 import argparse
 import os
-import sys
 import time
 from tqdm import tqdm
 from pathlib import Path
 
-sys.path.append("./")
-
 from lib.reconstructor import Reconstructor
 from lib import utils
 from lib.utils import cprint, Col, get_user_confirmation

scripts/remesh.py

@@ -4,7 +4,6 @@
 sys.path.append("./")
 
 from lib.remesher import remesh, save_mesh
-from lib.visualize_model import render_3d_model
 from lib.utils import cprint, Col
 from lib.led_map_3d import LEDMap3D
 
@@ -15,7 +14,7 @@
     parser.add_argument(
         "map_filename",
         type=str,
-        help="The CSV 3D map file generated by reconstruct.py",
+        help="The CSV 3D map file generated by marimapper.py",
     )
 
     parser.add_argument(
@@ -34,11 +33,10 @@
 
     led_map = LEDMap3D(args.map_filename)
 
-    if led_map is None:
+    if not led_map.valid:
         quit()
 
     mesh = remesh(led_map, args.detail)
 
     if not save_mesh(mesh, args.mesh_filename):
         cprint(f"Failed to save mesh to {args.mesh_filename}", format=Col.FAIL)
-    render_3d_model(led_map, mesh=mesh)

scripts/view_2d_map.py

@@ -16,7 +16,9 @@ def render_2d_model(led_map):
 
     for led_id in led_map.get_detections():
         col = colorsys.hsv_to_rgb(led_id / max_id, 0.5, 1)
-        pos = np.array((led_map.get_detection(led_id).u, led_map.get_detection(led_id).v))
+        pos = np.array(
+            (led_map.get_detection(led_id).u, led_map.get_detection(led_id).v)
+        )
         image_point = (pos * 640).astype(int)
         cv2.drawMarker(display, image_point, color=col)
         cv2.putText(