FMG + IMU

Capture
Research validation

Forearm sensing for multimodal research

System overview

Built around
your research.

Pair forearm pressure patterns associated with muscle deformation and inertial motion with first-person video. A head-mounted camera and two forearm bands keep the palms and fingertips free for natural interaction.

Pressure patterns and motion: FMG + IMU
Left and right forearm capture: Bilateral
First-person task context: Ego pairing

Configuration, lead time and delivery scope confirmed with your inquiry.

Explore in 3D

FMG + IMU / In three dimensions

Explore the system.

Loading 3D preview...

Conceptual exterior and layout only, not engineering CAD. Internal structures, materials and dimensions require project-specific confirmation.

Design & capability

Designed for the task.

Hands remain free

Forearm bands support natural contact without covering the palms or fingertips. They do not replace every tactile or hand-pose sensor.

Complement visual observations

A modular band combines inward-facing contact elements and an IMU. Forearm pressure changes are not measurements of fingertip-object force.

Test the value of each signal

Gesture, 3D hand-pose and fingertip-pressure prediction are research directions requiring independent labels, cross-user testing and robot task comparisons.

Specifications

Configuration details and measurement conditions.

Sensing architecture
Signals	FMG + IMU; no EMG channels
Form	Modular forearm band with inward-facing contact elements
Array	Demonstration material labels a 10 x 4 three-dimensional force-sensing array; measurement points, axes and independent channel count require confirmation
Frequency	120 Hz is stated in demonstration material; sampling, transmission refresh and algorithm output rates must be defined separately
IMU	Axes, range, noise and sampling rate to be confirmed
Wearable setup	Head-mounted stereo vision + bilateral forearm bands; single / paired package scope to be confirmed
Communication	Wireless protocol, bandwidth, range and SDK to be confirmed

Do not infer 40 or 120 data channels from the 10 x 4 label or the rendered module count.

Data and evaluation
Raw records	FMG outputs, inertial signals, timestamps, sequence numbers and quality flags; final schema requires interface confirmation
Derived outputs	Model estimates of gestures, hand pose or fingertip pressure; dimensions, rate and accuracy require validation
Reference measurements	Independent hand-pose, force or tactile sensors and annotations selected for the task
Cross-device alignment	Clock mapping and synchronization events with Ego, UMI or robots; verify the error separately
Wear metadata	Anonymous participant ID, hand, position, orientation, fit, resting baseline, date and re-wearing events
Export	Session / episode organization with raw data, calibration and derived outputs; protocol and SDK licenses to be confirmed

Engineering items to confirm
Sensing	Range, accuracy, sensitivity, drift, hysteresis, overload and temperature effects
Fit	Forearm circumference, module adjustment, preload, long-session comfort and skin-contact materials
Power / structure	Weight, battery, runtime, charging, protection and cleaning
Link	End-to-end latency, packet loss, coexistence, reconnect and device concurrency
Software	Operating systems, raw-data access, SDK, model licenses and user calibration
Acceptance	Independent users, cross-day re-wearing and matched task baselines

Delivery and validation

Agree on the complete configuration.

Research validation stage. Wearing concepts and demonstration videos are not test reports. FMG is not EMG or direct fingertip force ground truth. No unverified production, delivery or robot-performance claims are implied.

Freeze sensor and communication specifications before defining the schema.
Calibrate fit and signal quality for each wearing session.
Use independent reference sensors and store observations separately from predictions.
Compare models with and without FMG under the same task and training budget.

Start with your task.

Define the devices, data and evaluation your project needs.

FMG + IMU

Forearm sensing for multimodal research

Built around
your research.

Explore the system.