Interesting Engineering

Autonomous humanoids gain safer navigation with advanced 3D perception tech

RealSense demos safer humanoid navigation at NVIDIA GTC, using 3D vision and simulation to enable reliable real-world robot movement.
GitHub

cuVSLAM: CUDA-Accelerated Visual Odometry and Mapping

cuVSLAM is the library by NVIDIA, providing various Visual Tracking Camera modes and Simultaneous Localization and Mapping (SLAM) capabilities. Leveraging CUDA acceleration and a rich set of features, ...
IEEE

Robust Visual Odometry Using Rigidly-Bundled Arbitrarily-Arranged Multi-Cameras

Abstract: Making multi-camera visual SLAM systems easier to set up and more robust to the environment is attractive for vision robots. Existing monocular and binocular vision SLAM systems have narrow ...
IEEE

Fast Monocular Visual-Inertial Odometry With Point-Line Features and Vanishing Point Constraints

Abstract: Visual-inertial odometry (VIO) is widely used in autonomous driving, robots, and drones in global navigation satellite system (GNSS)-denied environments. Current VIO methods primarily rely ...
marktechpost

[Tutorial] Building a Visual Document Retrieval Pipeline with ColPali and Late Interaction Scoring

In this tutorial, we build an end-to-end visual document retrieval pipeline using ColPali. We focus on making the setup robust by resolving common dependency conflicts and ensuring the environment ...