Structured-Light 3D Cameras for Robot Vision: Selection Guide
A structured-light 3D camera must resolve the target geometry throughout a defined working volume, under the materials and lighting found in production. A field of view that contains the bin is not enough if point spacing, depth uncertainty, occlusion or reflection prevents the system from locating the grasp or inspection feature. Camera selection also affects mounting, calibration, cycle time and the compute stack. Give suppliers representative parts, scene images and the required output. Request point-cloud trials at the intended distance and orientation, with documented settings and pass criteria, before locking the camera model and software configuration.
How to shortlist
Working volume
Match near and far working distance, field of view and depth of field to the complete bin, pallet or inspection volume, including mounting tolerance and robot motion.
Required geometric detail
Define the smallest feature, edge, gap or pose difference that must be resolved. Compare point spacing and measured accuracy at the actual working distance.
Surface behavior
Test representative dark, reflective, translucent, textured and low-contrast parts. State whether surface treatment is allowed and which failure rate is acceptable.
Ambient light
Record illuminance, direct sunlight, flicker and changing shadows. Ask for tests under the worst expected light rather than relying on a maximum marketing value.
Acquisition and cycle time
Include exposure, multi-frame capture, transfer, point-cloud processing and robot settling in the cycle budget. Confirm whether moving objects must be frozen.
Occlusion and camera placement
Compare fixed, eye-in-hand and multi-camera layouts. Check blind regions, robot shadows, cable motion, collision clearance and whether more than one view is needed.
Calibration and coordinate stability
Define hand-eye or fixed-camera calibration, allowable drift, recalibration triggers, calibration target access and how coordinate transforms are validated.
Industrial and software integration
Check enclosure rating, temperature, vibration, trigger and synchronization, network bandwidth, data format, SDK platform, GPU or IPC needs and long-term software compatibility.
Matching catalog models
15 modelsSide-by-side specifications
| Model | Technology | Depth range | Depth accuracy | Resolution | Frame rate | Field of view | Interface | Quote | |
|---|---|---|---|---|---|---|---|---|---|
| RevopointTrackit SR | Structured light 3D scanning† | — | — | — | — | — | — | Inquiry | Quote |
| RevopointMIRACO Plus | Structured light 3D scanning† | — | — | — | — | — | — | Inquiry | Quote |
| RevopointPOP 4 | Structured light 3D scanning† | — | — | — | — | — | — | Inquiry | Quote |
| RevopointMetroY Ultra | Structured light 3D scanning† | — | — | — | — | — | — | Inquiry | Quote |
| RevopointMetroX Pro | Structured light 3D scanning† | — | — | — | — | — | — | Inquiry | Quote |
| DeptrumAurora 930 | Structured light (nano-photonic projector)† | — | — | — | — | — | — | Inquiry | Quote |
| PercipioFM855-E1 | Dot-pattern structured light† | — | — | — | — | — | Gigabit Ethernet† | Inquiry | Quote |
| PercipioFS820-E1 | Binocular speckle structured light† | — | — | — | — | — | Gigabit Ethernet† | Inquiry | Quote |
| PercipioFM851-E2 | Dot-pattern structured light† | — | — | — | — | — | Gigabit Ethernet† | Inquiry | Quote |
| Mech-MindLSR L | Structured laser light (Class 2)† | 1200-3000 mm† | 1 mm @3 m† | Depth 2048 x 1536 px; RGB 4000 x 3000 px† | — | 1200 x 1000 mm (minimum)† | GigE (C++, C#, Python, ROS SDK)† | Inquiry | Quote |
| Mech-MindPRO M | Structured light (LED)† | 1000-2000 mm† | 0.2 mm @2 m† | 1920 x 1200 px† | — | 800 x 450 mm (minimum)† | GigE (C++, C#, Python, ROS SDK)† | Inquiry | Quote |
| Mech-MindLOG M | Structured light (LED)† | 800-2000 mm† | 0.3 mm @2 m† | 1280 x 1024 px† | — | 520 x 390 mm (minimum)† | GigE (C++, C#, Python, ROS SDK)† | Inquiry | Quote |
† Manufacturer-reported, not yet independently verified.
Include these details for a comparable quote
- Representative parts, including the hardest colors, finishes and geometries
- Scene photographs and bin, pallet, fixture or inspection-station CAD
- Near and far working distance and required field of view at each plane
- Smallest feature, pose tolerance and acceptable measurement uncertainty
- Ambient-light range, direct-sun exposure and other optical interference
- Target acquisition time, full cycle time and whether objects are moving
- Fixed, eye-in-hand or multi-camera mounting preference and space limits
- Robot, PLC, trigger, synchronization, network and compute environment
- Required point-cloud, depth-map, RGB and confidence outputs
- Sample-test protocol, pass criteria, calibration and maintenance requirements