Maciej Wielgosz

In this post, we will explore the role of SLAM (Simultaneous Localization and Mapping) algorithms. SLAM is a computational technique that enables a device to map an unknown environment while concurrently determining its location. Utilizing LiDAR sensors, SLAM leverages high-resolution distance data to achieve precise localization and detailed environmental mapping. We discuss undistortion, feature point selection, frame matching and map expansion and touch upon additional processes.

In this post we'll delve into the world of point cloud data acquisition methods, comparing and contrasting the three main players: LiDAR, Photogrammetry, and Structured Light Scanning. Each method boasts unique strengths and weaknesses, making it crucial to understand their operational principles to choose the most suitable tool for your specific needs. We'll explore the mechanics behind each technique, uncovering their advantages, limitations, and ideal applications.

Point cloud data plays a crucial role in various 3D technologies. This blog post introduces this data format, explaining how it represents objects and scenes as collections of 3D points. We'll delve into the unique properties of point clouds, including their unordered nature and rich information content.

I'm excited to share an article about SegmentAnyTree, a sensor-agnostic deep-learning model capable of accurately identifying individual tree crowns from various laser scanning data sources, including airborne, terrestrial, and mobile platforms.

PedSynth, a large-scale synthetic dataset specifically designed for pedestrian intention prediction was presented at the 2023 IEEE Intelligent Transportation Systems Conference (ITSC).

Point2Tree goes beyond generic point cloud processing, employing a three-tiered approach to unlock new insights into forest structure and composition.

A sample dataset featuring pedestrians generated using a new, CARLA-based framework for generating datasets. It provides use cases for pedestrian detection, autoencoding, pose estimation, and pose lifting.

What is needed so that robots can work along with humans? Humans fear the unknown. Contemporary Artificial Intelligence (AI) systems have no way to directly incorporate any of the rules known to humans. Those two issues could be overcome if a way to extract, review, and update the rules governing the AI behavior existed.