Litcius/Paper detail

Lightweight Context-Aware Network Using Partial-Channel Transformation for Real-Time Semantic Segmentation

Min Shi, Shaowen Lin, Qingming Yi, Jian Weng, Aiwen Luo, Yicong Zhou

2024IEEE Transactions on Intelligent Transportation Systems86 citationsDOI

Abstract

Optimizing the computational efficiency of the artificial neural networks is crucial for resource-constrained platforms like autonomous driving systems. To address this challenge, we proposed a Lightweight Context-aware Network (LCNet) that accelerates semantic segmentation while maintaining a favorable trade-off between inference speed and segmentation accuracy in this paper. The proposed LCNet introduces a partial-channel transformation (PCT) strategy to minimize computing latency and hardware requirements of the basic unit. Within the PCT block, a three-branch context aggregation (TCA) module expands the feature receptive fields, capturing multiscale contextual information. Additionally, a dual-attention-guided decoder (DD) recovers spatial details and enhances pixel prediction accuracy. Extensive experiments on three benchmarks demonstrate the effectiveness and efficiency of the proposed LCNet model. Remarkably, a smaller model LCNet <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{3\_7}$</tex-math> </inline-formula> achieves 73.8% mIoU with only 0.51 million parameters, with an impressive inference speed of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim$</tex-math> </inline-formula> 142.5 fps and <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim$</tex-math> </inline-formula> 9 fps using a single RTX 3090 GPU and Jetson Xavier NX, respectively, on the Cityscapes test set at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1024\times 1024$</tex-math> </inline-formula> resolution. A more accurate version of the LCNet <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$_{3\_11}$</tex-math> </inline-formula> can achieve 75.8% mIoU with 0.74 million parameters at <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$\sim$</tex-math> </inline-formula> 117 fps inference speed on Cityscapes at the same resolution. Much faster inference speed can be achieved at smaller image resolutions. LCNet strikes a great balance between computational efficiency and prediction capability for mobile application scenarios. The code is available at https://github.com/lztjy/LCNet.

Topics & Concepts

Context (archaeology)NotationComputer scienceSegmentationInferenceTransformation (genetics)Artificial intelligenceArtificial neural networkAlgorithmTheoretical computer scienceMathematicsArithmeticBiochemistryChemistryPaleontologyGeneBiologyAdvanced Neural Network ApplicationsAdvanced Image and Video Retrieval TechniquesVideo Surveillance and Tracking Methods