# Semantic Segmentation for Autonomous Driving: Model Evaluation, Dataset Generation, Perspective Comparison, and Real-Time Capability

arXiv: https://arxiv.org/abs/2207.12939 · HTML: https://mlsanity.com/paper/2207.12939/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 31 references that are in the map.
- First posted on arXiv 26 Jul 2022.

## Builds on (most cited first, 13 of 13)

- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [The Cityscapes Dataset for Semantic Urban Scene Understanding](https://mlsanity.com/paper/1604.01685/) (2016)
- [Image Segmentation Using Deep Learning: A Survey](https://mlsanity.com/paper/2001.05566/) (2020)
- [BiSeNet: Bilateral Segmentation Network for Real-time Semantic Segmentation](https://mlsanity.com/paper/1808.00897/) (2018)
- [YOLACT: Real-time Instance Segmentation](https://mlsanity.com/paper/1904.02689/) (2019)
- [BiSeNet V2: Bilateral Network with Guided Aggregation for Real-time Semantic Segmentation](https://mlsanity.com/paper/2004.02147/) (2020)
- [CenterMask : Real-Time Anchor-Free Instance Segmentation](https://mlsanity.com/paper/1911.06667/) (2019)
- [BlendMask: Top-Down Meets Bottom-Up for Instance Segmentation](https://mlsanity.com/paper/2001.00309/) (2020)
- [In Defense of Pre-trained ImageNet Architectures for Real-time Semantic Segmentation of Road-driving Images](https://mlsanity.com/paper/1903.08469/) (2019)
- [YOLACT++: Better Real-time Instance Segmentation](https://mlsanity.com/paper/1912.06218/) (2019)
- [HarDNet: A Low Memory Traffic Network](https://mlsanity.com/paper/1909.00948/) (2019)
- [FasterSeg: Searching for Faster Real-time Semantic Segmentation](https://mlsanity.com/paper/1912.10917/) (2019)
- [Small Obstacle Avoidance Based on RGB-D Semantic Segmentation](https://mlsanity.com/paper/1908.11675/) (2019)

## Built on by (most cited first, 7 of 7)

- [One-shot Unsupervised Domain Adaptation with Personalized Diffusion Models](https://mlsanity.com/paper/2303.18080/) (2023)
- [A Survey of Vision Transformers in Autonomous Driving: Current Trends and Future Directions](https://mlsanity.com/paper/2403.07542/) (2024)
- [Flashy Backdoor: Real-world Environment Backdoor Attack on SNNs with DVS Cameras](https://mlsanity.com/paper/2411.03022/) (2024)
- [OVeNet: Offset Vector Network for Semantic Segmentation](https://mlsanity.com/paper/2303.14516/) (2023)
- [Photonic Accelerators for Image Segmentation in Autonomous Driving and Defect Detection](https://mlsanity.com/paper/2309.16783/) (2023)
- [Hyperspectral image segmentation with a machine learning model trained using quantum annealer](https://mlsanity.com/paper/2503.01400/) (2025)
- [DSER: Spectral Epipolar Representation for Efficient Light Field Depth Estimation](https://mlsanity.com/paper/2508.08900/) (2025)

## Abstract (arXiv)

Environmental perception is an important aspect within the field of autonomous vehicles that provides crucial information about the driving domain, including but not limited to identifying clear driving areas and surrounding obstacles. Semantic segmentation is a widely used perception method for self-driving cars that associates each pixel of an image with a predefined class. In this context, several segmentation models are evaluated regarding accuracy and efficiency. Experimental results on the generated dataset confirm that the segmentation model FasterSeg is fast enough to be used in realtime on lowpower computational (embedded) devices in self-driving cars. A simple method is also introduced to generate synthetic training data for the model. Moreover, the accuracy of the first-person perspective and the bird's eye view perspective are compared. For a 320 × 256 input in the first-person perspective, FasterSeg achieves 65.44,% mean Intersection over Union (mIoU), and for a 320 × 256 input from the bird's eye view perspective, FasterSeg achieves 64.08,% mIoU. Both perspectives achieve a frame rate of 247.11 Frames per Second (FPS) on the NVIDIA Jetson AGX Xavier. Lastly, the frame rate and the accuracy with respect to the arithmetic 16-bit Floating Point (FP16) and 32-bit Floating Point (FP32) of both perspectives are measured and compared on the target hardware.