
1 / 4 · Click to enlarge
AI Summary
A compact offline AI vision module that recognises five hand gestures and up to ten faces at once within 3 metres. It runs entirely on-device, so Arduino, Raspberry Pi and ESP32 projects gain touchless control without any cloud dependency.
Key Highlights
- Onboard edge AI processing keeps all detection local and private
- Recognises five hand gestures for touchless control
- Detects up to 10 faces or upper-body presences simultaneously
- 3-metre detection range with an 85-degree field of view
- Runs on 3.3 to 5 V with around 100 mA current draw
- Gravity PH2.0 connectors for quick Arduino, RPi and ESP32 wiring
Ideal For
- Makers building touchless control into Arduino or ESP32 projects
- Raspberry Pi developers adding presence detection to kiosks
- Smart home enthusiasts wanting private, cloud-free sensing
- Educators teaching edge AI and computer vision basics
- Product developers prototyping gesture-controlled interfaces
Buying Tip
Check that your controller board can supply around 100 mA at 3.3 to 5 V, and plan the mounting angle so faces and hands fall inside the 85-degree, 3-metre detection cone.
Elecvora expert review
Our verdict
A neatly packaged edge AI sensor that removes the cloud from gesture and face detection, which is exactly what privacy-sensitive and latency-sensitive projects need. The Gravity connector and broad board support make it quick to prototype with.
AI-assisted analysis based on the manufacturer's specifications. Not a hands-on lab test.
Pros
- Detection runs fully on-device with no cloud dependency
- Handles five gestures and up to ten faces simultaneously
- Wide 3.3 to 5 V input suits most maker boards
- Gravity PH2.0 cabling keeps wiring simple
- Compact 42 x 32 mm, 6 g form factor
Cons
- Detection range is limited to about 3 metres
- No enclosure or mounting bracket is included
- Requires a separate controller board and power source
Best for
Makers and developers adding private, touchless gesture and presence sensing to Arduino, Raspberry Pi or ESP32 projects
Not ideal for
Long-range surveillance or outdoor detection beyond 3 metres
Specifications
| Weight (g) | 06 |
|---|---|
| Operating Voltage (V) | 3.3 – 5 |
| Current consumption | 100 |
| Operating Current (A) | 100 |
| Length (mm) | 42 |
| Width (mm) | 32 |
| Shipping Weight | 0.016 kg |
| Shipping Dimensions | 6 × 5.5 × 2.5 cm |
What's in the box
- 1 x Gravity Offline Edge AI Gesture and Face Detection Sensor
- 1 x Gravity PH2.0-4P connecting line
- 1 x double-ended PH2.0-4P cable
Frequently Asked Questions
Does this sensor need an internet connection?
Which boards can I use it with?
What voltage and current does it require?
How many gestures and faces can it detect?
What is the detection range and viewing angle?
What comes in the package?
Overview
The DFRobot Gravity Offline Edge AI Gesture and Face Detection Sensor is a small vision module built for touchless interaction. Instead of streaming images to a server, it runs its AI models on an onboard processor, so gesture and face detection happen in real time on the device itself. That makes it a good fit for projects where privacy, latency or an unreliable network connection matter.
The module recognises five hand gestures and can detect up to ten faces or upper-body presences at the same time, at a range of up to 3 metres with an 85-degree field of view. It measures just 42 x 32 mm and weighs about 6 grams, so it fits easily into compact enclosures and panel-mounted builds.
Key Features
- Offline edge AI: detection runs locally with no cloud dependency.
- Multi-person sensing: up to 10 faces or upper-body presences at once.
- Gesture control: five hand gestures for touchless commands.
- Wide coverage: 3-metre range with an 85-degree field of view.
- Low power: approximately 100 mA at 3.3 to 5 V.
- Simple wiring: Gravity PH2.0 connectors and included cables.
Specifications Explained
Operating voltage (3.3 to 5 V): the module works with both 3.3 V and 5 V logic boards, so it can share a supply with most Arduino, Raspberry Pi and ESP32 setups.
Current consumption (about 100 mA): modest enough for USB-powered boards and small battery packs, though you should still allow headroom in your power budget.
Detection range (3 m) and field of view (85 degrees): the sensor covers a room-scale area in front of it, which suits doorway, kiosk and panel installations. Position it so the area of interest sits inside that cone.
Size and weight (42 x 32 mm, 6 g): light enough to mount on a bracket or behind a display panel without extra support.
Compatibility and Requirements
The sensor is designed to work with Arduino, Raspberry Pi and ESP32 boards, and it supports graphical programming environments such as MakeCode. A suitable controller board, a 3.3 to 5 V power source and the included Gravity PH2.0 cable are all you need to get started.
Typical Uses
- Contactless control of information screens in public areas
- Gesture-driven stage and performance effects
- Touchless control of home appliances
- Music playback control without touching a device
- Occupancy and presence sensing for smart spaces
What's in the Box
- 1 x Gravity Offline Edge AI Gesture and Face Detection Sensor
- 1 x Gravity PH2.0-4P connecting line
- 1 x double-ended PH2.0-4P cable

Quick specs
- Weight (g)
- 06
- Operating Voltage (V)
- 3.3 – 5
- Current consumption
- 100
- Operating Current (A)
- 100
- Length (mm)
- 42
- Width (mm)
- 32


















