diff --git a/code/backend_reconaissance/Dockerfile b/code/backend_reconaissance/Dockerfile
index 7de5fb7..b3535ee 100644
--- a/code/backend_reconaissance/Dockerfile
+++ b/code/backend_reconaissance/Dockerfile
@@ -5,7 +5,11 @@ ENV PYTHONDONTWRITEBYTECODE=1
 #Afficher les logs directement dans le terminal
 ENV PYTHONUNBUFFERED=1
 
-# Installation des dépendances
+#Installation des dépendances de opencv (TODO: supprimer si plus besoin)
+RUN apt-get update
+RUN apt-get install ffmpeg libsm6 libxext6  -y
+
+# Installation des dépendances python
 COPY requirements.txt .
 RUN python -m pip install -r requirements.txt
 
diff --git a/code/backend_reconaissance/__pycache__/hands.cpython-310.pyc b/code/backend_reconaissance/__pycache__/hands.cpython-310.pyc
new file mode 100644
index 0000000..ade5451
Binary files /dev/null and b/code/backend_reconaissance/__pycache__/hands.cpython-310.pyc differ
diff --git a/code/backend_reconaissance/hands.py b/code/backend_reconaissance/hands.py
new file mode 100644
index 0000000..ddf8dbf
--- /dev/null
+++ b/code/backend_reconaissance/hands.py
@@ -0,0 +1,45 @@
+import cv2
+import mediapipe as mp
+mp_drawing = mp.solutions.drawing_utils
+mp_drawing_styles = mp.solutions.drawing_styles
+mp_hands = mp.solutions.hands
+
+# For webcam input:
+# cap = cv2.VideoCapture("dev/v4l/by-path/pci-0000:04:00.3-usb-0:4:1.0-video-index0")
+cap = cv2.VideoCapture(0)
+hands = mp_hands.Hands(
+    model_complexity=0,
+    min_detection_confidence=0.5,
+    min_tracking_confidence=0.5)
+
+def frame():
+  if cap.isOpened():
+    success, image = cap.read()
+    if not success:
+      print("Ignoring empty camera frame.")
+      # If loading a video, use 'break' instead of 'continue'.
+      return
+
+    # To improve performance, optionally mark the image as not writeable to
+    # pass by reference.
+    image.flags.writeable = False
+    image = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)
+    results = hands.process(image)
+
+    # Draw the hand annotations on the image.
+    image.flags.writeable = True
+    image = cv2.cvtColor(image, cv2.COLOR_RGB2BGR)
+    if results.multi_hand_landmarks:
+      for hand_landmarks in results.multi_hand_landmarks:
+        mp_drawing.draw_landmarks(
+            image,
+            hand_landmarks,
+            mp_hands.HAND_CONNECTIONS,
+            mp_drawing_styles.get_default_hand_landmarks_style(),
+            mp_drawing_styles.get_default_hand_connections_style())
+    # Flip the image horizontally for a selfie-view display.
+    # cv2.imshow('MediaPipe Hands', cv2.flip(image, 1))
+    if cv2.waitKey(5) & 0xFF == 27:
+      return
+# cap.release()
+
diff --git a/code/backend_reconaissance/main.py b/code/backend_reconaissance/main.py
index 41ca251..7f83423 100644
--- a/code/backend_reconaissance/main.py
+++ b/code/backend_reconaissance/main.py
@@ -3,6 +3,7 @@ import json
 import websockets
 import random
 import os
+import hands
 
 class WebsocketServer:
     def __init__(self,getEffects,port=os.getenv("PORT"),host=os.getenv("HOST")) -> None:
@@ -18,6 +19,7 @@ class WebsocketServer:
     async def handler(self,websocket):
         while True:
             messages = self.getEffects()
+            hands.frame()
             await websocket.send(json.dumps(messages))
             await asyncio.sleep(1/30)
 
diff --git a/code/backend_reconaissance/requirements.txt b/code/backend_reconaissance/requirements.txt
index d992242..2376504 100644
--- a/code/backend_reconaissance/requirements.txt
+++ b/code/backend_reconaissance/requirements.txt
@@ -1,2 +1,4 @@
 websockets
-requests
\ No newline at end of file
+requests
+opencv-python
+mediapipe
\ No newline at end of file
diff --git a/code/docker-compose.yaml b/code/docker-compose.yaml
index 295f1da..e195644 100644
--- a/code/docker-compose.yaml
+++ b/code/docker-compose.yaml
@@ -79,6 +79,8 @@ services:
     build: ./backend_reconaissance
     container_name: backend_reconaissance
     restart: always
+    devices:
+      - /dev/video3:/dev/video0
     environment:
       - PORT=5000
       - HOST=backend_reconaissance
diff --git a/docs/dupliquer_camera.md b/docs/dupliquer_camera.md
index 96164a1..3dc4500 100644
--- a/docs/dupliquer_camera.md
+++ b/docs/dupliquer_camera.md
@@ -3,6 +3,8 @@
     * Télécharger : `git clone https://github.com/umlaeute/v4l2loopback.git`
     * Installer avec : `make` puis `sudo make install`
     * activer le module : `sudo modprobe v4l2loopback devices=2`
+        * Erreur possible : opperation not permitted : il faut désactiver secure boot
+    * OU `apt update && apt install v4l2loopback-dkms v4l2loopback-utils`
 * Faire looper la camera /dev/video0 sur les autres
     * installer ffmpeg : `sudo apt get install ffmpeg`
     * activer le loopback : `ffmpeg -f video4linux2 -i /dev/video0 -codec copy -f v4l2 /dev/video1 -codec copy -f v4l2 /dev/video2`