mlsanityData to 7 Oct 2026

Multi-encoder ConvNeXt Network with Smooth Attentional Feature Fusion for Multispectral Semantic Segmentation

arXiv 2602.10137 ↗First posted 8 Feb 2026cs.CV cs.AIcited by 3

Not on the map. Why?

Abstract (arXiv)

This work proposes MeCSAFNet, a multi-branch encoder-decoder architecture for land cover segmentation in multispectral imagery. The model separately processes visible and non-visible channels through dual ConvNeXt encoders, followed by individual decoders that reconstruct spatial information. A dedicated fusion decoder integrates intermediate features at multiple scales, combining fine spatial cues with high-level spectral representations. The feature fusion is further enhanced with CBAM attention, and the ASAU activation function contributes to stable and efficient optimization. The model is designed to process different spectral configurations, including a 4-channel (4c) input combining RGB and NIR bands, as well as a 6-channel (6c) input incorporating NDVI and NDWI indices. Experiments on the Five-Billion-Pixels (FBP) and Potsdam datasets demonstrate significant performance gains. On FBP, MeCSAFNet-base (6c) surpasses U-Net (4c) by +19.21%, U-Net (6c) by +14.72%, SegFormer (4c) by +19.62%, and SegFormer (6c) by +14.74% in mIoU. On Potsdam, MeCSAFNet-large (4c) improves over DeepLabV3+ (4c) by +6.48%, DeepLabV3+ (6c) by +5.85%, SegFormer (4c) by +9.11%, and SegFormer (6c) by +4.80% in mIoU. The model also achieves consistent gains over several recent state-of-the-art approaches. Moreover, compact variants of MeCSAFNet deliver notable performance with lower training time and reduced inference cost, supporting their deployment in resource-constrained environments. Model code is available at: https://github.com/Leo-Thomas/mecsafnet

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 17 of its 93 references that are in the map. First posted on arXiv 8 Feb 2026.

Builds on

17 of 93 references.

References · 17

  1. 2020A Simple Framework for Contrastive Learning of Visual Representationscited by 11,979
  2. 2016Feature Pyramid Networks for Object Detectioncited by 5,021
  3. 2022A ConvNet for the 2020scited by 3,545
  4. 2018CBAM: Convolutional Block Attention Modulecited by 2,463
  5. 2016Xception: Deep Learning with Depthwise Separable Convolutionscited by 2,320
  6. 2016Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Networkcited by 1,199
  7. 2018ADVENT: Adversarial Entropy Minimization for Domain Adaptation in Semantic Segmentationcited by 623
  8. 2023SpectralGPT: Spectral Remote Sensing Foundation Modelcited by 250
  9. 2018Land-Cover Classification with High-Resolution Remote Sensing Images Using Transferable Deep Modelscited by 118
  10. 2018Contextual Hourglass Network for Semantic Segmentation of High Resolution Aerial Imagerycited by 42
  11. 2022Enabling Country-Scale Land Cover Mapping with Meter-Resolution Satellite Imagerycited by 31
  12. 2020Multi-stage Attention ResU-Net for Semantic Segmentation of Fine-Resolution Remote Sensing Imagescited by 26
5 more
  1. 2022Self-supervised remote sensing feature learning: Learning Paradigms, Challenges, and Future Workscited by 24
  2. 2021A review of machine learning in processing remote sensing data for mineral explorationcited by 16
  3. 2022False: False Negative Samples Aware Contrastive Learning for Semantic Segmentation of High-Resolution Remote Sensing Imagecited by 8
  4. 2021An Attention-Fused Network for Semantic Segmentation of Very-High-Resolution Remote Sensing Imagerycited by 7
  5. 2022Finding Differences Between Transformers and ConvNets Using Counterfactual Simulation Testingcited by 6

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026A Parameter-efficient Convolutional Approach for Camouflaged Weed Detection in Multispectral Aerial Imagerycited by 1
  2. 2026StrokeNeXt: A Siamese-encoder Approach for Brain Stroke Classification in Computed Tomography Imagerycited by 0
  3. 2026Lightweight Polyp Segmentation via a Gain-Aware Prediction-Space Recursive Controllercited by 0