Enabling Real-Time Semantic Segmentation for an Automotive Application

0 X
Lower Latency
0 X
Model Size Reduction
0 X
Memory Footprint Reduction


Automotive company



Use case

Semantic Segmentation


Advanced Driver Assistance Systems (ADAS) technology powered by deep learning models is transforming the mobility and transportation industry. However, AI developers are still facing an uphill battle when trying to transition from the lab to real-world deployments. A common barrier to deployment is the inability to achieve high accuracy and real-time inference performance on edge devices. Both factors are absolutely mission-critical for ensuring not only the application’s usability but also safety for the users.

Leading automotive manufacturers and ADAS providers use Deci to boost their models’ performance and deploy multiple models on edge devices.

The Challenge

An automotive company running a U-Net based segmentation model on a NVIDIA Jetson Xavier NX struggled to achieve the target latency in production.

The Solution

Using Deci’s AutoNAC engine, a faster and smaller model was generated. Latency was reduced by 2.1x, model size was reduced by 3x, and memory footprint was reduced by 3x – all while maintaining the original accuracy.

The Results

Lower Latency
Model Size Reduction
Memory Footprint Reduction

Why Deci?

Achieve Real-Time Inference
on Edge Devices

Improve latency and throughput, and reduce model size by up to 5X while maintaining the model’s accuracy.

Process More Video Streams
on Less Devices

Maximize hardware utilization and cost-efficiently scale your solution at the edge.

Reduce Development
Effort and Risks

Simplify development with automated tools that guarantee success.

How It Works:

Deci Platform

Foundation or Custom Models

Choose an ultra performant model or generate a custom one.



Neural Architecture Search Engine

On Prem


Dataset Analyzer

Train or

Use Deci’s library & custom recipe to train on-prem.

On Prem


PyTorch Training Library

Optimize & Run

Apply acceleration techniques. 
Run self-hosted inference anywhere.

On Prem


Optimization & Inference Engine SDK

Talk to Our Experts

Build Better Models Faster with Deci’s Deep Learning Development Platform

Tell us about your use case, needs, goals, and the obstacles in your way. We’ll show you how you can use the Deci platform to overcome them.

Book a Demo

Add Your Heading Text Here
					from transformers import AutoFeatureExtractor, AutoModelForImageClassification

extractor = AutoFeatureExtractor.from_pretrained("microsoft/resnet-50")

model = AutoModelForImageClassification.from_pretrained("microsoft/resnet-50")