TwinLiteNet+: An enhanced multi-task segmentation model for autonomous driving

Semantic segmentation is a fundamental perception task in autonomous driving, particularly for identifying drivable areas and lane markings to enable safe navigation. However, most state-of-the-art (SOTA) models are computationally intensive and unsuitable for real-time deployment on resource-constrained embedded devices. In this paper, we introduce TwinLiteNet${}^{+}$, an enhanced multi-task segmentation model designed for real-time drivable area and lane segmentation with high efficiency. TwinLiteNet${}^{+}$ employs a hybrid encoder architecture that integrates stride-based dilated convolutions and depthwise separable dilated convolutions, balancing representational capacity and computational cost. To improve task-specific decoding, we propose two lightweight upsampling modules-Upper Convolution Block (UCB) and Upper Simple Block (USB)-alongside a Partial Class Activation Attention (PCAA) mechanism that enhances segmentation precision. The model is available in four configurations, ranging from the ultra-compact TwinLiteNet${}_{\text{Nano}}^{+}$ (34K parameters) to the high-performance TwinLiteNet${}_{\text{Large}}^{+}$ (1.94M parameters). On the BDD100K dataset (Yu et al. (2020)), TwinLiteNet${}_{\text{Large}}^{+}$ achieves 92.9% mIoU for drivable area segmentation and 34.2% IoU for lane segmentation-surpassing existing state-of-the-art models while requiring 11$\times$ fewer floating-point operations (FLOPs) for computation. The results compared with other models are shown in Fig. 1. Extensive evaluations on embedded devices demonstrate superior inference speed, quantization robustness (INT8/FP16), and energy efficiency, validating TwinLiteNet${}^{+}$ as a compelling solution for real-world autonomous driving systems. Code is available at https://github.com/chequanghuy/TwinLiteNetPlus.