Specifications
Models
Femto Bolt (USB-C, host processing), Femto Mega (USB/Ethernet, Jetson Nano onboard), Femto Mega I (IP65, M12 Ethernet, Jetson Nano onboard)
Depth Technology
Indirect Time of Flight (iToF) at 850 nm
Depth Range
0.25 m to 5.46 m depending on depth mode
Depth Resolution / FPS
Up to 1024 x 1024 @ 15 fps (WFOV) or 640 x 576 @ 30 fps (NFOV) in stream mode; Femto Mega I snapshot mode is 1024 x 1024 @ 10 fps / 640 x 576 @ 10 fps
Depth Field of View
H 120° x V 120° wide field of view, or H 75° x V 65° narrow field of view
Depth Accuracy
Typical systematic error < 11 mm + 0.1% of distance
Depth Precision
Random error standard deviation <= 17 mm
RGB Camera
Up to 3840 x 2160 (4K) with HDR — 30 fps on Femto Bolt, 25 fps on Femto Mega, 25 fps stream / 3 fps snapshot on Femto Mega I
RGB Field of View
H 80° x V 51°, with colour-to-depth registration
Onboard Processing
NVIDIA Jetson Nano module performs the in-camera depth engine on Femto Mega and Femto Mega I; Femto Bolt offloads to the host
IMU
6 degrees of freedom, with pose and skeleton tracking support
Multi-Camera Sync
8-pin hardware synchronisation trigger system; any camera or external sensor can act as primary
Femto Bolt Physical
115 x 40 x 65 mm, 348 g, USB 3.2 Type-C Gen 1, 4.35 W average
Femto Mega Physical
115 x 40 x 145 mm, 560 g, USB 3.0 Type-C or Ethernet, PoE+ / DC 12 V 2 A, average < 13 W
Femto Mega I Physical
180 x 50 x 110 mm, 1080 g, IP65, M12 X-coding 8-pin Ethernet with PoE+ or DC 12-24 V, average < 11 W, 8 x M5 mounting
Operating Environment
Femto Bolt and Femto Mega 10 °C to 25 °C indoor; Femto Mega I 0 °C to 45 °C indoor/semi-outdoor, 8-90% RH non-condensing
SDK
Orbbec SDK and Orbbec Industry SDK, plus an Azure Kinect Sensor SDK (K4A) API wrapper for drop-in migration from AKDK
Data Output
Point cloud, depth map, IR and RGB frames
Overview
The Orbbec Femto family is a line of indirect time-of-flight depth and RGB cameras that reuse the Microsoft ToF technology found in HoloLens and the Azure Kinect Developer Kit. All three models share the same depth camera modes and performance as the AKDK, which makes them a drop-in replacement path for the large installed base of Azure Kinect applications now that the developer kit is discontinued — Orbbec ships an Azure Kinect Sensor SDK API wrapper alongside the native Orbbec SDK so existing code can be repointed with minimal change.
The three models are separated by connection and compute, not by depth quality. Femto Bolt is the compact host-processed device: 115 x 40 x 65 mm, 348 g and only 4.35 W average over a single USB 3.2 Type-C cable, adding HDR support on the colour camera. Femto Mega integrates an NVIDIA Jetson Nano module so the depth engine runs inside the camera and processed depth, IR and 4K RGB stream over USB or Ethernet, with PoE+ available for single-cable installation. Femto Mega I is the industrial model: an IP65 housing with an M12 X-coding industrial Ethernet connector, PoE+ or 12-24 VDC input, a wider industrial temperature range and 8 x M5 mounting points.
Depth coverage is a 120° x 120° wide field of view at up to 1024 x 1024, or a 75° x 65° narrow field of view that trades coverage for range up to 5.46 m. Accuracy is specified as typical systematic error below 11 mm plus 0.1% of distance, with a random error standard deviation of 17 mm or better. A 6DoF IMU and an 8-pin synchronisation bus allow several cameras to be fused into a wider capture volume, and body-tracking and point-cloud output are available through the SDK.
Key Benefits
Azure Kinect DK replacement path with identical depth modes and a K4A API wrapper for migration. In-camera depth processing on Femto Mega and Mega I via an integrated NVIDIA Jetson Nano, removing host CPU load. Industrially packaged option — Femto Mega I adds IP65 sealing, an M12 X-coding Ethernet port, PoE+ and a 0-45 °C range. Widest field of view in class at 120° x 120°. 4K RGB with HDR and colour-to-depth registration for fused colour point clouds. Single-cable installs with PoE+ power and data on one Ethernet run. Multi-camera synchronisation via an 8-pin trigger bus for volumetric capture and multi-view robotics.
Applications
Robotic picking, depalletisation and loading/unloading with precise object pose. Pallet and package dimensioning in logistics and shipping. Human pose and motion analysis for rehabilitation, elderly care and sports biomechanics. Volumetric video reconstruction for AR/XR and interactive broadcast. Quality control and 3D surface measurement in manufacturing. Vision-guided palletising and machine tending. Medical patient positioning and guided workout applications.
Request a Quote — ORBBEC FEMTO BOLT / FEMTO MEGA / FEMTO MEGA I — TOF DEPTH CAMERAS WITH NVIDIA JETSON
QS Compute — global B2B supply of AI computing hardware, edge AI systems and accelerators. Volume pricing, 15-day sample lead time.
Get Your Quote →