Realtime semantic segmentation

Semantic segmentation is a computer vision technique that labels each pixel in an image according to its category, such as person, background, road, or car. It's used in applications like background removal and autonomous driving, where detailed scene understanding is crucial. Real-time performance is important, especially in fast-paced environments, to ensure accurate and timely decisions based on the latest visual data.

At NeuralSpike we developed a cutting-edge real-time semantic segmentation system that operates efficiently on both web browsers (via WebAssembly) and edge devices like System On Module (SOM) or even Microcontroller Units (MCU). For instance on Mediatek Genio or NVIDIA Jetson. Our solution outperforms the popular MediaPipe model in terms of speed and accuracy. For instance, on MediaTek Genio we achieved an impressive 90 FPS vs 80 FPS MediaPipes. Our model is also more accurate, and can segment finer objects like fingers.

To improve our model even further we used model distillation technique to transfer knowledge from a complex but slower model. Depending on the deployment target we could improve the speed by applying other techniques such as model quantization.

Our segmentation model is a great illustration of our model training pipeline. The example model segments people from the background, can be deployed on MediaTek Genio or NVIDIA Jetson. But our training process is very flexible. Depending on our clients needs, we can train models with different segmentation classes (e.g. vehicles). It can be also deployed on a different hardware platform. In such cases we can apply our architecture search solution to find the most suitable model for selected hardware.

Request your AI consultation

Consent*