3D RealSense Depth Camera

Buy in Store

Overview

The 3D RealSense Depth Camera is a high-performance visual perception device available in D435i, D405, and D405CB models. It supports face analysis, augmented reality, object tracking, 3D scanning, and is specially optimized for embodied intelligence development.

Key features:

• Three models covering near-to-far range and precision needs
• High-precision depth, RGB, and infrared output (D435i also provides IMU data)
• Optimized for embodied AI: autonomous navigation, object recognition, manipulation
• Optional adaptation for XLeRobot and SO-ARM101 robot platforms, plug-and-play

---

Model Comparison

Model	Range	Best For
D435i	Mid-to-long range	Mobile robot navigation, environment 3D reconstruction
D405	Short range, high precision	Robot arm grasping, close-range object recognition
D405CB	Short range (D405 enhanced)	Complex/low-light environments, higher precision

Specifications

Category	Spec
Models	D435i / D405 / D405CB
Core Functions	Face analysis, AR, object tracking, 3D scanning, embodied AI perception
Compatible Platforms	XLeRobot / SO-ARM101 (optional)
Output	Depth, RGB, IR, IMU (D435i)
Use Cases	Robotics, AI research, 3D reconstruction, industrial inspection, AR/VR, embodied AI

---

Quick Start

1. Install Driver

# Ubuntu 22.04 (X86 / Jetson)
pip install pyrealsense2

2. Verify Device

rs-enumerate-devices

You should see your RealSense camera and its model.

3. Basic Example

import pyrealsense2 as rs
import numpy as np
import cv2

pipeline = rs.pipeline()
config = rs.config()
config.enable_stream(rs.stream.depth, 640, 480, rs.format.z16, 30)
config.enable_stream(rs.stream.color, 640, 480, rs.format.bgr8, 30)

pipeline.start(config)

try:
    while True:
        frames = pipeline.wait_for_frames()
        depth = frames.get_depth_frame()
        color = frames.get_color_frame()
        if not depth or not color:
            continue
        depth_image = np.asanyarray(depth.get_data())
        color_image = np.asanyarray(color.get_data())
        cv2.imshow('Color', color_image)
        cv2.imshow('Depth', depth_image * 80)
        if cv2.waitKey(1) & 0xFF == ord('q'):
            break
finally:
    pipeline.stop()
    cv2.destroyAllWindows()

4. LeRobot Integration

For SO-ARM101 / XLeRobot projects:

# Find camera ID
python -m lerobot.find_cameras realsense

# Enable RealSense during teleoperation
lerobot-teleoperate \\
  --robot.cameras='{ front: {type: realsense} }' \\
  ...

---

Applications

Scenario	Description
Face Analysis	Recognition, expression, attribute analysis
Augmented Reality	Overlay, spatial localization, 3D registration
Object Tracking	Detection, tracking, counting
3D Scanning	Model reconstruction, volume measurement, dimension check
Embodied AI	Environment perception, obstacle avoidance, manipulation

---

FAQ

Q: How to choose a model?

A:

• Mobile robot navigation / reconstruction → D435i (mid-range, with IMU)
• Robot arm grasping / close-range → D405 (compact, high precision)
• Low-light / complex environments → D405CB (D405 enhanced)

Q: Does it support Jetson?

A: Yes. pyrealsense2 installs directly on Jetson and is compatible with the SO-ARM101 LeRobot workflow.

---

Support

• 📧 Email: support@juxitech.com
• 🌐 Website: www.juxitech.com
• 💬 Report Issues