Spatial Reasoning from Sensor Data
About This Course
Unlock the potential of spatial sensing and reasoning technology in this course, designed for professionals seeking to master the art of interpreting sensor data for a variety of cutting-edge applications. Once reserved for specialised enterprise and industrial use, such as urban planning and robotic navigation, these technologies have now expanded into a plethora of markets, including automotive, consumer goods, and industrial sectors.
This course provides a strong theoretical foundation while emphasizing hands-on skills through practical activities. Participants will work on real-world case studies, such as image-based localization for mobile robots and object detection from Light Detection and Ranging (LiDAR) data in autonomous driving systems. Using Python, they will build deep learning models to analyze sensor data, including multi-view images, Red-Green-Blue and Depth (RGB-D) images, and point cloud data.
What You'll Learn
- Acquire the fundamentals of spatial sensing and spatial reasoning technology, including scene representation, scanning and mapping, spatial reasoning using feature-based approaches and the state-of-the-art machine learning based approaches.
- Apply critical analysis, evaluation and synthesis to a wide range of spatial sensing and spatial reasoning problems.
- Build spatial sensing and spatial reasoning systems in diverse areas such as games, robotics, and augmented reality.
Entry Requirements
Participants should have intermediate skills in Python programming (e.g., Numpy, Pandas), and/or OpenCV programming (e.g., able to apply filtering and transformation on the image).
Experienced in using Jupyter Notebooks, Google Colab, and well-versed in package installation.