Augment your data with RealSim
Apply realistic textures onto your existing RGB datasets to increase variation for perception training.
Hover over the image to see RealSim results!
Unannotated real-world images are used to fine-tune RealSim to learn how the world looks.
Generate data in simulation to obtain synthetic images, and feed this into RealSim
RealSim applies real-world textures on to the synthetic dataset to further stimulate the model during training.
Better results with fewer resources.
RealSim enables perception engineers to obtain more variety in their data with fewer real-world annotated images. This reduces cost without compromising model performance.
“We’ve seen models train comparatively using RealSim data as they would using annotated images logged from the real world.”
We’re developing RealSim in automotive/AD, off-road domains such as mining, and even manufacturing.
RealSim has a substantial amount of pre-trained knowledge. It only requires a limited volume of real-world images to fine-tune to a new domain.
Virtual Worlds with Infinite Possibilities.