Face-Tracking-using-ESP32-CAM/index.html at main · MatsRobot/Face-Tracking-using-ESP32-CAM · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
<!DOCTYPE html>
<html lang="en">
<head>
    <meta charset="UTF-8">
    <meta name="viewport" content="width=device-width, initial-scale=1.0">
    <link rel="icon" type="image/png" href="https://matsrobot.github.io/Favicon-192.png" sizes="192x192">

    <title>Real-Time Face Tracking: ESP32-CAM & PCA9685 | MatsRobot</title>
    <meta name="description" content="A standalone real-time face tracking system using ESP32-CAM and PCA9685. Features onboard Haar Cascade detection, I2C servo control, and autonomous PID logic—no Wi-Fi required.">

    <link rel="canonical" href="https://github.com/MatsRobot/Face-Tracking-using-ESP32-CAM" />

    <style>
        :root {
            --gh-green: #238636;
            --gh-green-hover: #2ea44f;
            --gh-dark: #24292e;
            --gh-blue: #0366d6;
            --gh-border: #eaecef;
            --gh-bg-gray: #f6f8fa;
            --note-bg: #fffbdd;
            --note-border: #d4a017;
        }

        body {
            font-family: -apple-system, BlinkMacSystemFont, "Segoe UI", Helvetica, Arial, sans-serif;
            line-height: 1.6;
            color: var(--gh-dark);
            max-width: 950px;
            margin: 0 auto;
            padding: 2rem;
        }

        .hero-section {
            display: flex;
            gap: 2.5rem;
            margin-bottom: 2rem;
            align-items: center;
        }

        .hero-text { flex: 3; }

        .image-sidebar {
            flex: 2;
            border: 1px solid var(--gh-border);
            border-radius: 8px;
            padding: 15px;
            text-align: center;
            background: #fff;
            box-shadow: 0 2px 5px rgba(0,0,0,0.05);
        }

        .image-sidebar img {
            max-width: 100%;
            height: auto;
            border-radius: 4px;
        }

        .caption {
            font-size: 0.85rem;
            color: #586069;
            margin-top: 10px;
            display: block;
        }

        h1 { font-size: 1.8rem; margin-top: 0; }
        h2 { color: var(--gh-blue); border-bottom: 1px solid var(--gh-border); padding-bottom: 0.3em; margin-top: 2rem; }
        h3 { font-size: 1.1rem; margin-bottom: 0.5rem; }

        .note {
            background-color: var(--note-bg);
            padding: 1.25rem;
            border-left: 5px solid var(--note-border);
            margin: 1.5rem 0;
            border-radius: 4px;
        }

        .btn-download {
            background-color: var(--gh-green);
            color: white;
            padding: 12px 28px;
            text-decoration: none;
            border-radius: 6px;
            display: inline-flex;
            align-items: center;
            gap: 18px;
            font-weight: 600;
            transition: background 0.2s, transform 0.1s;
            border: 1px solid rgba(27, 31, 35, 0.15);
            box-shadow: 0 1px 0 rgba(27, 31, 35, 0.1);
        }

        .btn-download:hover { background-color: var(--gh-green-hover); }

        .btn-text-container {
            display: flex;
            flex-direction: column;
            line-height: 1.1;
        }

        .btn-main {
            font-size: 1.35rem;
            text-transform: uppercase;
            letter-spacing: 0.5px;
            margin-bottom: 4px;
        }

        .btn-sub-lines {
            font-size: 0.75rem;
            font-weight: 400;
            opacity: 0.9;
        }

        /* Mobile optimization for tables */
        .table-container {
            overflow-x: auto;
            margin: 1rem 0;
        }

        .tech-table {
            width: 100%;
            border-collapse: collapse;
            min-width: 500px; /* Ensures table doesn't squish too much on mobile */
        }

        .tech-table th, .tech-table td {
            padding: 12px;
            border: 1px solid var(--gh-border);
        }

        .tech-table th { background-color: var(--gh-bg-gray); text-align: left; }

        footer {
            margin-top: 4rem;
            padding: 2rem 0;
            border-top: 1px solid var(--gh-border);
            text-align: center;
            color: #6a737d;
            font-size: 0.9rem;
        }

        @media (max-width: 768px) {
            .hero-section { flex-direction: column; }
            .hero-text { order: 2; }
            .image-sidebar { order: 1; width: 100%; }
            .btn-download { width: 100%; justify-content: center; box-sizing: border-box; }
        }
    </style>
</head>
<body>

    <div class="hero-section">
        <div class="hero-text">
            <h1>🤖 FaceTracker: Standalone ESP32-CAM System</h1>
            <p>A self-contained computer vision solution that tracks human faces in real-time. This project transforms the <strong>ESP32-CAM</strong> into an active robotic sensor that independently calculates coordinates—completely independent of external apps.</p>

            <div class="note">
                <strong>No Middleman:</strong> This system processes detection onboard and acts immediately, functioning without any Wi-Fi connection.
            </div>

            <a href="https://github.com/MatsRobot/Face-Tracking-using-ESP32-CAM" class="btn-download">
                <svg width="32" height="32" viewBox="0 0 24 24" fill="white">
                    <path d="M19 9h-4V3H9v6H5l7 7 7-7zM5 18v2h14v-2H5z"/>
                </svg>
                <div class="btn-text-container">
                    <span class="btn-main">Get Source Code</span>
                    <div class="btn-sub-lines">
                        <span>Full Build Instructions & ESP-WHO Logic</span>
                    </div>
                </div>
            </a>
        </div>

        <div class="image-sidebar">
            <img src="https://raw.githubusercontent.com/MatsRobot/Face-Tracking-using-ESP32-CAM/main/FaceTracker.gif" alt="ESP32-CAM and PCA9685 Wiring Diagram for Robotics Face Tracking">
            <span class="caption">Hardware Configuration</span>
        </div>
    </div>

    <h2>How It Works: The Logic Flow</h2>
    <p>The system operates in a continuous feedback loop to maintain target alignment:</p>
    <ul>
        <li><strong>Capture:</strong> The OV2640 sensor grabs a frame at QVGA resolution (320 x 240 pixels).</li>
        <li><strong>Detection:</strong> The system scans for face landmarks internally using the <strong>ESP-WHO framework</strong>.</li>
        <li><strong>Displacement Calculation:</strong> It determines the <strong>centroid coordinate</strong> distance from the frame center point (160, 120).</li>
        <li><strong>Actuation:</strong> PWM adjustment commands are sent via the <strong>400kHz I2C bus</strong> to the PCA9685 driver to re-center the camera.</li>
    </ul>

    <h2>Technical Control Loop & Logic</h2>
    <p>The software calculates the positional offset required to drive the Pan/Tilt motors with high precision:</p>
    <ul>
        <li><strong>Pan Offset:</strong> Calculated by subtracting the target's current X-coordinate from the frame center.</li>
        <li><strong>Tilt Offset:</strong> Calculated by subtracting the target's current Y-coordinate from the frame center.</li>
        <li><strong>Hardware Acceleration:</strong> Offloading 12-bit PWM processing to the PCA9685 prevents CPU bottlenecks and eliminates servo jitter common in software-timed pulses.</li>
    </ul>

    <h2>Technical Specifications</h2>
    <div class="table-container">
        <table class="tech-table">
            <thead>
                <tr><th>Component</th><th>Role & Specification</th></tr>
            </thead>
            <tbody>
                <tr><td><strong>Microcontroller</strong></td><td>ESP32-CAM (AI-Thinker) @ 240MHz</td></tr>
                <tr><td><strong>Servo Driver</strong></td><td>PCA9685 (12-bit PWM / I2C Address: 0x40)</td></tr>
                <tr><td><strong>Protocol</strong></td><td>I2C Bus (SDA/SCL) shared by SSD1306 & PCA9685</td></tr>
                <tr><td><strong>Power Management</strong></td><td>Isolated 5V rail for SG90/MG90S Micro Servos</td></tr>
                <tr><td><strong>Detection Engine</strong></td><td>ESP-WHO Framework (MTMN Neural Network)</td></tr>
            </tbody>
        </table>
    </div>

    <h2>🗺️ Future Roadmap</h2>
    <ul>
        <li><strong>S3-Series Migration:</strong> Moving to the <strong>ESP32-S3 CAM</strong> for higher FPS, while tackling new pinout challenges.</li>
        <li><strong>PID Control:</strong> Implementing smoother derivative logic for fluid tracking.</li>
        <li><strong>Target Prioritization:</strong> Smart logic to follow the largest subject in crowded frames.</li>
    </ul>

    <footer>
        <p>© 2026 MatsRobot | Part of the Open Source Robotics Initiative</p>
        <p><small>Copyright (c) 2026 | Licensed under the <a href="https://github.com/MatsRobot/matsrobot.github.io/blob/main/LICENSE" style="color: #6a737d;">MIT License</a></small></p>
    </footer>

</body>
</html>