Stamp CAM0308/CAM3660 Camera
The Stamp camera add-on is available with either a GC0308 sensor (CAM0308) or an OV3660 sensor (CAM3660).
The current UiFlow2 firmware supports both
variants on the StampS3Bat host and automatically detects when
camera.init is called.
Support the following products:
Supported sensors
The sensor driver identifies the camera over SCCB (the I2C-compatible camera control bus). The supported sensor variants are:
Product |
Sensor |
SCCB address |
Maximum resolution |
Supported output formats |
|---|---|---|---|---|
CAM0308 |
GC0308 |
|
640 x 480 (VGA) |
RGB565, YUV422, grayscale |
CAM3660 |
OV3660 |
|
2048 x 1536 (QXGA) |
RGB565, YUV422, JPEG |
The camera wiring and power-control configuration in the current firmware is specific to the StampS3Bat controller. Attach the camera add-on before initializing the camera.
UiFlow2 Example
Capture and draw on an image
Open the cam0308_image_example.m5f2 project in UiFlow2. The example initializes the camera at QVGA/RGB565, captures one frame, draws text and shapes on it, and prints the image dimensions and format.
Wi-Fi video streaming with QR recognition
Open the cam0308_video_streaming_example.m5f2 project in UiFlow2. Replace the Wi-Fi SSID and password in the project, then open the printed device IP address on port 8080 in a browser. The stream includes QR-code recognition results when a QR code is visible.
MicroPython Example
Capture and draw on an image
1# SPDX-FileCopyrightText: 2026 M5Stack Technology CO LTD
2#
3# SPDX-License-Identifier: MIT
4
5import os, sys, io
6import M5
7from M5 import *
8import camera
9import image
10
11
12stamp_addon_cam0308_0 = None
13
14
15cam_img = None
16
17
18def setup():
19 global stamp_addon_cam0308_0, cam_img
20
21 M5.begin()
22 camera.init(pixformat=camera.RGB565, framesize=camera.FRAME_QVGA)
23 cam_img = camera.snapshot()
24 cam_img.draw_string(2, 2, str("cam0308"), color=0x3366FF, scale=1)
25 cam_img.draw_rectangle(2, 2, 316, 236, color=0x33CC00, thickness=1, fill=False)
26 cam_img.draw_line(0, 0, 319, 239, color=0xFF0000, thickness=1)
27 cam_img.draw_circle(160, 120, 50, color=0xFFCC00, thickness=1, fill=False)
28 print(cam_img.width())
29 print(cam_img.height())
30 print(cam_img.format())
31 print(cam_img.size())
32
33
34def loop():
35 global stamp_addon_cam0308_0, cam_img
36 M5.update()
37
38
39if __name__ == "__main__":
40 try:
41 setup()
42 while True:
43 loop()
44 except (Exception, KeyboardInterrupt) as e:
45 try:
46 from utility import print_error_msg
47
48 print_error_msg(e)
49 except ImportError:
50 print("please update to latest firmware")
Wi-Fi video streaming with QR recognition
Set ssid and pwd in the example before running it. After the board
connects to Wi-Fi, visit http://<device-ip>:8080/ to view the MJPEG stream.
1# SPDX-FileCopyrightText: 2026 M5Stack Technology CO LTD
2#
3# SPDX-License-Identifier: MIT
4
5import os, sys, io
6import M5
7from M5 import *
8import camera
9from easysocket import EasyTCPServer
10import socket
11import network
12import code_scanner
13import time
14import image
15import jpg
16
17
18tcps = None
19wlan_sta = None
20stamp_addon_cam0308_0 = None
21
22
23stream_info = None
24client = None
25request = None
26ssid = None
27pwd = None
28cam_img = None
29code_result = None
30frame = None
31
32
33def setup():
34 global \
35 tcps, \
36 wlan_sta, \
37 stamp_addon_cam0308_0, \
38 stream_info, \
39 client, \
40 request, \
41 ssid, \
42 pwd, \
43 cam_img, \
44 code_result, \
45 frame
46
47 M5.begin()
48 camera.init(pixformat=camera.RGB565, framesize=camera.FRAME_QVGA)
49 camera.set_hmirror(False)
50 camera.set_vflip(False)
51 wlan_sta = network.WLAN(network.STA_IF)
52 wlan_sta.active(True)
53 ssid = "your-wifi-ssid"
54 pwd = "your-wifi-pwd"
55 print((str("Connecting to Wi-Fi: ") + str(ssid)))
56 wlan_sta.connect(ssid, pwd)
57 while not (wlan_sta.isconnected()):
58 time.sleep_ms(100)
59 tcps = socket.socket(socket.AF_INET, socket.SOCK_STREAM)
60 tcps.setsockopt(socket.SOL_SOCKET, socket.SO_REUSEADDR, 1)
61 tcps.bind(("0.0.0.0", 8080))
62 tcps.listen(1)
63 print(
64 (str("Open http://") + str((str((wlan_sta.ifconfig()[0])) + str(":8080/ in a browser"))))
65 )
66
67
68def loop():
69 global \
70 tcps, \
71 wlan_sta, \
72 stamp_addon_cam0308_0, \
73 stream_info, \
74 client, \
75 request, \
76 ssid, \
77 pwd, \
78 cam_img, \
79 code_result, \
80 frame
81 M5.update()
82 try:
83 M5.update()
84 stream_info = tcps.accept()
85 client = stream_info[0]
86 request = client.recv(1024)
87 client.send(
88 (
89 str(
90 (
91 str(
92 (
93 str(
94 (
95 str(
96 (
97 str(
98 (
99 str(
100 (
101 str("HTTP/1.1 200 OK")
102 + str((chr(13)))
103 )
104 )
105 + str((chr(10)))
106 )
107 )
108 + str(
109 "Content-Type: multipart/x-mixed-replace; boundary=cam0308"
110 )
111 )
112 )
113 + str((chr(13)))
114 )
115 )
116 + str((chr(10)))
117 )
118 )
119 + str((chr(13)))
120 )
121 )
122 + str((chr(10)))
123 )
124 )
125 while True:
126 cam_img = camera.snapshot()
127 code_result = code_scanner.find_qrcodes(cam_img)
128 if code_result:
129 print(code_result.payload())
130 cam_img.draw_string(8, 8, str(code_result.payload()), color=0x0000FF, scale=1)
131 frame = jpg.encode(cam_img, 80)
132 client.send(
133 (
134 str(
135 (
136 str(
137 (
138 str(
139 (
140 str(
141 (
142 str(
143 (
144 str(
145 (
146 str(
147 (
148 str(
149 (
150 str((chr(13)))
151 + str(
152 (chr(10))
153 )
154 )
155 )
156 + str("--cam0308")
157 )
158 )
159 + str((chr(13)))
160 )
161 )
162 + str((chr(10)))
163 )
164 )
165 + str("Content-Type: image/jpeg")
166 )
167 )
168 + str((chr(13)))
169 )
170 )
171 + str((chr(10)))
172 )
173 )
174 + str((chr(13)))
175 )
176 )
177 + str((chr(10)))
178 )
179 )
180 client.send(frame.bytearray())
181 except:
182 try:
183 client.close()
184 except:
185 pass
186
187 tcps.close()
188 stream_info = tcps.accept()
189
190
191if __name__ == "__main__":
192 try:
193 setup()
194 while True:
195 loop()
196 except (Exception, KeyboardInterrupt) as e:
197 try:
198 from utility import print_error_msg
199
200 print_error_msg(e)
201 except ImportError:
202 print("please update to latest firmware")
API
The camera add-on uses the shared camera module. See
camera for initialization, capture, frame-size, pixel
format, mirror, and flip APIs.
The image returned by camera.snapshot() uses the shared image module;
see image for image drawing and inspection methods.



