Winning at CVPR 2019: Mapillary Tops Two Computer Vision Benchmarking Challenges
At CVPR this year, Mapillary won two computer vision benchmarking challenges. We will always keep pushing the boundaries of what is possible in computer vision, and it is our award-winning models that allow us to produce the highest quality map data possible.
Introducing the Mapillary Traffic Sign Dataset for Teaching Machines to Understand Traffic Signs Globally
Today we’re releasing the Mapillary Traffic Sign Dataset, the world’s most diverse publicly available dataset of traffic sign annotations on street-level imagery that will help improve traffic safety and navigation everywhere. Covering different regions, weather and light conditions, camera sensors, and viewpoints, it enables developing high-performing traffic sign recognition models in both academic and commercial research.
Training Machines to Attain a 3D Understanding of Objects from Single, 2D Images
We sit down with Peter Kontschieder, the Director of Research at Mapillary, to talk about “Disentangling Monocular 3D Object Detection”, the latest academic paper to be published by Mapillary’s Research team. Peter tells us about how 3D object detections made in single 2D images have the ability to improve mapmaking and push down the cost of autonomous vehicles, and how the team unveiled a fundamental flaw in the metric used by the most dominant benchmarking dataset in this area.
Introducing Seamless Scene Segmentation: Allowing Machines to Understand Street Scenes Better by Turning Two Models into One
Today we’re announcing that Mapillary will publish four papers at CVPR this year. In this post, we’re looking at the paper named Seamless Scene Segmentation, which, as a world-first, rolls out a new computer vision model that slashes up to 20% computing powers when teaching machines to distinguish between people, cars, and map data like traffic signs, together with its overall environment.
Full Speed Ahead: How Toyota Research Institute is Accelerating its Machine Learning Algorithms with Mapillary
Toyota Research Institute (TRI) is focused on developing state-of-the-art machine learning algorithms for autonomous driving to realize safe and accessible mobility for the future. In this blog post, Jie Li, Research Scientist at TRI, outlines how TRI utilizes the Mapillary Vistas Dataset as a benchmark for driving scene understanding algorithms providing geometric and semantic diversity at scale.
Doing way more with Less: Catching up with the Mapillary Research Team
Since 2016 when we opened our AI lab in Graz, the research team has been busy publishing papers, winning benchmarking competitions, and developing the building blocks that power Mapillary. Now we are celebrating the opening of a brand new Graz lab and looking back at how it all came together.
Analyzing Parking Signs at Scale: How Mapillary is Working with Amazon Rekognition to Help US Cities End Their Parking Troubles
Managing parking infrastructure is a billion-dollar problem for cities all across the US. There has been no easy way for cities and Departments of Transportation to access parking sign data, resulting in poor decisions around parking infrastructure and planning. Today, Mapillary and Amazon Rekognition are introducing a scalable way to help US cities get a handle of their parking infrastructure.
Building the Tools to Show Us the Way: How Mapillary is Ramping up Traffic Sign Recognition Globally
We’re releasing an update to Mapillary’s traffic sign recognition, featuring wider support of traffic sign classes globally, improved recognition accuracy, and traffic sign taxonomy.
Massive Memory Savings for Training Modern Deep Learning Architectures
Mapillary Research has developed a novel approach to training recognition models to handle up to 50% more training data than before in every single learning iteration. With this technology, we can improve over the winning semantic segmentation method of this year’s Large-Scale Scene Understanding Workshop on the challenging Mapillary Vistas Dataset, setting a new state of the art.
Human in the Loop: Perfecting AI Algorithms
Machine learning needs human input. By creating a loop where human feedback is provided to the output of AI detection algorithms, we can significantly improve the accuracy of the models and the resulting map data.
Map Data in the Era of Autonomous Driving
The development of autonomous driving sets high requirements to map data. Next to using advanced equipment to collect map data, collaborative mapping combined with computer vision is a lower-cost, faster, more scalable approach.
How to Make Time Travel Happen
The Time Travel feature on Mapillary is great for observing how places change in time. Here's an insight into how it works and what you can do to get more matches between images.
More Accurate Map Data: Improving 3D Reconstruction with Semantic Understanding
Reconstructing a 3D world from 2D images is not as straightforward for a computer as it is for humans due to the fact that some objects are moving around in the real world. Understanding an image scene through semantic segmentation improves the 3D reconstruction, resulting in more accurate map data and better navigation in the image viewer.
Towards Global Traffic Sign Recognition
We are taking a big step towards recognizing traffic signs all over the world by adding support for more than 500 traffic signs globally, together with an appearance-based taxonomy for traffic signs. This is the beginning of our journey of recognizing every road sign in the world, no matter where it is.
Building the MapillaryJS Navigation Graph
In MapillaryJS 2.0 we completely changed the way we retrieve data and build the navigation graph to improve performance. Here is how it works.