Summary of ICCV 2021 Workshop on Map-Based Localization for Autonomous Driving In case you couldn’t attend ICCV, let us give you a quick update on what you missed from — of course our favorite — the workshop on AV mapping. The workshop was held on Oct 11th and chaired by…

What you need to know about paths, transforms (TF) and other settings In this article, we will talk about how an autonomous vehicle can know its own location. But first, let’s start with a simple example. Introduction If you are driving a car in an unfamiliar place, you usually rely on…

Adapting LOAM to work with REMOVERT and running it on an example from nuScenes dataset Introduction Creating a map from from LiDAR scans is nowadays a standard process. It’s known as a process called point cloud registration. It’s done by stitching together several point clouds that are recorded in each instance…

Image segmentation is one of the fundamental steps toward scene understanding of machines. By doing image segmentation, machines transit from the abstract image categorization toward more grounded pixel-level classification, in which each pixel is labeled by considering its local neighborhood, image context, scene composition as well as available low-level (pixel…

Here we show visual slam for monocular videos and how a consistent map is obtained via loop closure mechanism. For self-driving cars to localize themselves in a global environment, accurate geometric maps are needed. Self-driving cars need to know in which lane they are driving but also how far away…

Self-supervised learning is all the rage these days for machine learning. Whether its models like GPT-3 for natural language processing or data augmentation for computer vision, everyone is trying to get something for free from their data without paying for expensive things like labels. In this article, we’ll go over…