Triple-View Knowledge Distillation for Semi-Supervised Semantic Segmentation

doi:10.1109/TMM.2025.3639923

Triple-View Knowledge Distillation for Semi-Supervised Semantic Segmentation

2025-12-03

0

PRE

AI

DOI：10.1109/TMM.2025.3639923

Original

Original request for help

Save

Abstract

En 中文

To alleviate the expensive human labeling problem, semi-supervised semantic segmentation utilizes a few labeled images along with an abundance of unlabeled images to predict the pixel-level label maps with the same size. Previous methods often rely on co-training with two convolutional networks with the same architecture but different initialization, which fails to capture sufficiently diverse features. This limitation motivates us to employ tri-training and design a triple-view encoder to utilize encoders with different architectures to derive diverse features, while leveraging knowledge distillation to capture complementary semantics among these encoders. Moreover, existing approaches simply concatenate features from both encoder and decoder, and the simple concatenation requires a large memory cost. This inspires us to present a dual-frequency decoder that selects those important features by projecting the spatial-domain features into the frequency domain, where a dual-frequency channel attention mechanism is applied to evaluate the feature importance. Therefore, we propose a Triple-view Knowledge Distillation framework, termed TriKD, for semi-supervised semantic segmentation. It comprises the triple-view encoder and the dual-frequency decoder. Extensive experiments conducted on two benchmarks, i.e., Pascal VOC 2012 and Cityscapes, validate the superiority of our method, achieving a satisfying tradeoff between precision and inference speed.

Keywords:

Semantic segmentation

semi-supervised learning

knowledge distillation

channel-wise attention

triple-view encoder

Triple-View Knowledge Distillation for Semi-Supervised Semantic Segmentation

Abstract

Journal

Organization