Intel RealSense Stereoscopic Depth Cameras

By L. Keselman, John Iselin Woodfill, Anders Grunnet-Jepsen, Achintya Bhowmik. Computer Vision and Pattern Recognition, 2017

404 citations recorded by the source corpus.

We present a comprehensive overview of the stereoscopic Intel RealSense RGBD imaging systems. We discuss these systems' mode-of-operation, functional behavior and include models of their expected performance, shortcomings, and limitations. We provide information about the systems' optical characteristics, their correlation algorithms, and how these properties can affect different applications, including 3D reconstruction and gesture recognition. Our discussion covers the Intel RealSense R200 and the Intel RealSense D400 (formally RS400).

Subfields: Computer Vision and Pattern Recognition, Media Technology

View Paper: https://www.semanticscholar.org/paper/2c6801308889f8f4b36856fd540841f9ddc74942

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Media Technology

Citations in the corpus

Swarm of micro flying robots in the wild (2022, methodology): “We use a grayscale and depth camera Intel Realsense D430 (71, 72) and an IMU from the FCU.”
An Efficient Sampling-Based Method for Online Informative Path Planning in Unknown Environments (2019, methodology): “Commonly used 3D sensors, such as the Microsoft Kinect or Intel Realsense, have uncertainties that scale approximately quadratically with depth z [18], [19].”
BAD SLAM: Bundle Adjusted Direct RGB-D SLAM (2019, methodology): “[31, 33, 48, 71], we use active stereo: By doing stereo depth estimation on two IR cameras, the stereo algorithm takes advantage of both the active illumination and ambient infrared light.”
Learning Depth with Convolutional Spatial Propagation Network (2018, methodology): “1b), where a pair of calibrated images, commonly from a stereo camera [27], are used as inputs.”
Learning high-speed flight in the wild (2021, methodology): “Because off-the-shelf depth sensors, such as the Intel RealSense 435 camera used on our physical platform, compute depth from stereo images using similar principles (40), this strategy ensures that the characteristics of the input align between simulation and the real world.”
Domain Randomization and Generative Models for Robotic Grasping (2017, methodology): “At test time, the depth input was produced by an Intel RealSense D435 [16].”
Skeleton-based action recognition via spatial and temporal transformer networks (2020, background): “Recent advances in 3D depth cameras such as Microsoft Kinect (Zhang (2012)) and Intel RealSense (Keselman et al. (2017)) sensors, and advanced human pose estimation algorithms (Cao et al. (2019)) made it possible to estimate 3D skeleton coordinates quickly and accurately with cheap devices.”