Real-Time Semantic Segmentation with Optimized RetinaNet Architectures for Embedded Automotive Systems
Quick Answer
This paper shows that The Opt-RetinaSeg architecture optimizes semantic segmentation for embedded automotive systems, achieving 73.9% mIoU at 70.4 FPS on NVIDIA Jetson Xavier NX.
Quick Take
This model offers a 7.4x speedup and 4x size reduction compared to ResNet-50 with minimal accuracy loss, making it suitable for real-time ADAS applications.
Key Points
- Opt-RetinaSeg replaces ResNet-50 with a lightweight feature extractor.
- Achieves 73.9% mIoU and 70.4 FPS on Cityscapes and BDD100K datasets.
- Utilizes structured channel pruning and post-training INT8 quantization.
- 4x reduction in model size with less than 3% accuracy degradation.
- Demonstrates viability for real-time segmentation in automotive perception.
Paper Resources
Source Excerpt
Real-time perception is a foundational requirement for advanced driver assistance systems (ADAS) and autonomous vehicles, yet embedded automotive platforms impose severe constraints on compute, memory, and power. This paper presents an optimized semantic segmentation architecture derived from the RetinaNet detection framework, adapted for dense pixel-wise prediction and tailored for deployment on resource-constrained embedded hardware. The proposed architecture, termed Opt-RetinaSeg, replaces th
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →ProMoE-FL: Prototype-conditioned Mixture of Experts for Multimodal Federated Learning with Missing Modalities
ProMoE-FL introduces a Prototype-conditioned Mixture-of-Experts framework for multimodal federated learning, effectively addressing missing modalities. It outperforms existing methods on four chest X-ray datasets, demonstrating superior feature synthesis capabilities in both homogeneous and heterogeneous settings.
