Our takeaways from AutoSens Brussels 2021 - Anyverse

Anyverse™ takeaways from AutoSens Brussels 2021


Our takeaways from AutoSens Brussels 2021

Last week, we had the pleasure to attend the AutoSens Brussels 2021 as exhibitors. Being back at face-to-face events felt great!
First of all, we would like to thank AutoSens for conducting a safe event and reuniting a vast number of leading engineers and technical experts from across the ADAS and autonomous vehicle world, as well as a special mention to the venue. Placing a technology conference for autonomous vehicles in such a place surrounded by automobile history while you are helping shape its future is definitely a great idea.

Having said that, these are our key takeaways:

Sensor simulation: the hottest topic

Sensors are not commodities, they are an essential part of a perception system and need to be carefully crafted for the operation and the environment. Especially if we take into account the current shift to data-centered AI that makes high-fidelity sensor simulation more critical for synthetic data pipelines.

A future where we can understand how the different sensors or different configurations affect the performance of AI perception systems is not far. An accurate sensor simulation is one of the key enabling technologies of that future.

Save time & costs - Simulate sensors!

Physically-based sensor simulation to train, test, and validate your computer perception deep learning model

In-cabin sensing gains traction

As the industry shifts its focus from exterior to interior, from driver to occupancy, from basic monitoring to advanced monitoring, and from safety as the main feature, In-Cabin Sensing systems have been emerging at an unprecedented pace.

Interesting conferences and talks took place where we were able to discuss enabling technologies, different in-cabin infrastructures, different sensing methods, and sensor fusion, but today, camera-based interior monitoring systems combined with deep learning artificial intelligence are the only top-performing technology to provide the system with the information it needs.

The need for more reliable sensing systems

There is no longer any doubt about the importance of camera-based sensing systems for Level 2+ application. They are the “default” system configuration, which puts the sensors at the heart of the sensing structure and confirms the need and vital importance of feeding our AI model with accurate data that matches exactly our system’s design, adapted to our specific sensor and configurations, and ultimately, achieve a reliable and robust sensing system.

Improve perception robustness: vital to achieving safer autonomous vehicles and ADAS

Vehicles must operate safely under any type of circumstance or weather condition, and even when great progress has been made to enable ADAS and autonomous vehicles to become safer, it is not enough.

Numerous reports show a lack of robustness in harsh scenarios, such as rain, snow, or fog, where they experience darkness and poor weather. We can find several advanced machine learning approaches to improve computer vision accuracy under all scenarios, but there is certainly a lot of work to do on safety matters. Sensors, like high-resolution LiDAR and thermal cameras, have a huge role to play, being able to simulate them could mean a huge advantage to reach the most demanding robustness goals.

About Anyverse™

Anyverse™ helps you continuously improve your deep learning perception models to reduce your system’s time to market applying new software 2.0 processes. Our synthetic data production platform allows us to provide high-fidelity accurate and balanced datasets. Along with a data-driven iterative process, we can help you reach the required model performance.

With Anyverse™ you can accurately simulate any camera sensor and help you decide which one will perform better with your perception system. No more complex and expensive experiments with real devices, thanks to our state-of-the-art photometric pipeline.

Need to know more?

Come visit our website anyverse.ai anytime, or our Linkedin, and Twitter social media profiles.

Scroll to Top

Let's talk about synthetic data!