Multi-Resolution End-to-End Deep Neural Network for Optimizing Latency-Accuracy Tradeoff in Autonomous Driving

📰 ArXiv cs.AI

Learn to optimize latency-accuracy tradeoffs in autonomous driving using a multi-resolution end-to-end deep neural network

advanced Published 29 May 2026

Action Steps

Design a multi-resolution end-to-end deep neural network to handle varying scene contexts and compute availability
Implement a latency-aware optimization algorithm to adjust the network configuration in real-time
Train the network using a dataset of diverse driving scenarios to improve prediction quality and reduce latency
Evaluate the network's performance using metrics such as mean average precision and latency
Fine-tune the network's hyperparameters to achieve the optimal latency-accuracy tradeoff

Who Needs to Know This

Autonomous driving engineers and researchers can benefit from this approach to improve the safety and efficiency of their systems

Key Insight

💡 A single fixed-resolution model is not optimal for autonomous driving; a multi-resolution approach can adapt to varying scene contexts and compute availability

Key Takeaways

Learn to optimize latency-accuracy tradeoffs in autonomous driving using a multi-resolution end-to-end deep neural network

Full Article

Title: Multi-Resolution End-to-End Deep Neural Network for Optimizing Latency-Accuracy Tradeoff in Autonomous Driving

Abstract:
arXiv:2605.29138v1 Announce Type: cross Abstract: Latency-accuracy tradeoffs are fundamental in real-time applications of deep neural networks (DNNs) for cyber-physical systems. In autonomous driving, in particular, safety depends on both prediction quality and the end-to-end delay from sensing to actuation. We observe that (1) when latency is accounted for, the latency-optimal network configuration varies with scene context and compute availability; and (2) a single fixed-resolution model becom

Read full paper → ← Back to Reads