# Exploiting Low-level Representations for Ultra-Fast Road Segmentation

arXiv: https://arxiv.org/abs/2402.02430 · HTML: https://mlsanity.com/paper/2402.02430/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 68 references that are in the map.
- First posted on arXiv 4 Feb 2024.

## Builds on (most cited first, 30 of 30)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Fully Convolutional Networks for Semantic Segmentation](https://mlsanity.com/paper/1411.4038/) (2014)
- [MobileNetV2: Inverted Residuals and Linear Bottlenecks](https://mlsanity.com/paper/1801.04381/) (2018)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [DeepLab: Semantic Image Segmentation with Deep Convolutional Nets, Atrous Convolution, and Fully Connected CRFs](https://mlsanity.com/paper/1606.00915/) (2016)
- [Pyramid Scene Parsing Network](https://mlsanity.com/paper/1612.01105/) (2016)
- [SegNet: A Deep Convolutional Encoder-Decoder Architecture for Image Segmentation](https://mlsanity.com/paper/1511.00561/) (2015)
- [Panoptic Segmentation](https://mlsanity.com/paper/1801.00868/) (2018)
- [BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation](https://mlsanity.com/paper/1808.00897/) (2018)
- [MobileViT: Light-weight, General-purpose, and Mobile-friendly Vision Transformer](https://mlsanity.com/paper/2110.02178/) (2021)
- [Training Region-based Object Detectors with Online Hard Example Mining](https://mlsanity.com/paper/1604.03540/) (2016)
- [BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation](https://mlsanity.com/paper/2004.02147/) (2020)
- [Fast-SCNN: Fast Semantic Segmentation Network](https://mlsanity.com/paper/1902.04502/) (2019)
- [TANet: Robust 3D Object Detection from Point Clouds with Triple Attention](https://mlsanity.com/paper/1912.05163/) (2019)
- [MultiNet: Real-time Joint Semantic Reasoning for Autonomous Driving](https://mlsanity.com/paper/1612.07695/) (2016)
- [Deep Dual-resolution Networks for Real-time and Accurate Semantic Segmentation of Road Scenes](https://mlsanity.com/paper/2101.06085/) (2021)
- [ContextNet: Exploring Context and Detail for Semantic Segmentation in Real-time](https://mlsanity.com/paper/1805.04554/) (2018)
- [Real-Time High-Performance Semantic Image Segmentation of Urban Street Scenes](https://mlsanity.com/paper/2003.08736/) (2020)
- [Learning Collision-Free Space Detection from Stereo Images: Homography Matrix Brings Better Data Augmentation](https://mlsanity.com/paper/2012.07890/) (2020)
- [Occlusion-shared and Feature-separated Network for Occlusion Relationship Reasoning](https://mlsanity.com/paper/1908.05898/) (2019)
- [A Novel Multi-layer Framework for Tiny Obstacle Discovery](https://mlsanity.com/paper/1904.10161/) (2019)
- [Real-time Semantic Segmentation with Context Aggregation Network](https://mlsanity.com/paper/2011.00993/) (2020)
- [Video Text Tracking With a Spatio-Temporal Complementary Model](https://mlsanity.com/paper/2111.04987/) (2021)
- [Tiny Obstacle Discovery by Occlusion-Aware Multilayer Regression](https://mlsanity.com/paper/2111.09204/) (2021)
- [Boundary-induced and scene-aggregated network for monocular depth prediction](https://mlsanity.com/paper/2102.13258/) (2021)
- [Indoor Obstacle Discovery on Reflective Ground via Monocular Camera](https://mlsanity.com/paper/2401.01445/) (2024)
- [Embedding Structured Contour and Location Prior in Siamesed Fully Convolutional Networks for Road Detection](https://mlsanity.com/paper/1905.01575/) (2019)
- [RoadNet-RT: High Throughput CNN Architecture and SoC Design for Real-Time Road Segmentation](https://mlsanity.com/paper/2006.07644/) (2020)
- [Vision-Based Road Detection using Contextual Blocks](https://mlsanity.com/paper/1509.01122/) (2015)

## Built on by (most cited first, 4 of 4)

- [SeaS: Few-shot Industrial Anomaly Image Generation with Separation and Sharing Fine-tuning](https://mlsanity.com/paper/2410.14987/) (2024)
- [AnomalyNCD: Towards Novel Anomaly Class Discovery in Industrial Scenarios](https://mlsanity.com/paper/2410.14379/) (2024)
- [EventSSEG: Event-driven Self-Supervised Segmentation with Probabilistic Attention](https://mlsanity.com/paper/2508.14856/) (2025)
- [Image-Intrinsic Priors for Integrated Circuit Defect Detection and Novel Class Discovery via Self-Supervised Learning](https://mlsanity.com/paper/2511.03120/) (2025)

## Abstract (arXiv)

Achieving real-time and accuracy on embedded platforms has always been the pursuit of road segmentation methods. To this end, they have proposed many lightweight networks. However, they ignore the fact that roads are "stuff" (background or environmental elements) rather than "things" (specific identifiable objects), which inspires us to explore the feasibility of representing roads with low-level instead of high-level features. Surprisingly, we find that the primary stage of mainstream network models is sufficient to represent most pixels of the road for segmentation. Motivated by this, we propose a Low-level Feature Dominated Road Segmentation network (LFD-RoadSeg). Specifically, LFD-RoadSeg employs a bilateral structure. The spatial detail branch is firstly designed to extract low-level feature representation for the road by the first stage of ResNet-18. To suppress texture-less regions mistaken as the road in the low-level feature, the context semantic branch is then designed to extract the context feature in a fast manner. To this end, in the second branch, we asymmetrically downsample the input image and design an aggregation module to achieve comparable receptive fields to the third stage of ResNet-18 but with less time consumption. Finally, to segment the road from the low-level feature, a selective fusion module is proposed to calculate pixel-wise attention between the low-level representation and context feature, and suppress the non-road low-level response by this attention. On KITTI-Road, LFD-RoadSeg achieves a maximum F1-measure (MaxF) of 95.21% and an average precision of 93.71%, while reaching 238 FPS on a single TITAN Xp and 54 FPS on a Jetson TX2, all with a compact model size of just 936k parameters. The source code is available at https://github.com/zhouhuan-hust/LFD-RoadSeg.