Research map / Image generation / Editing and style transfer
HiFC-GAN: Hierarchical Feature-Constrained GAN for Optical-to-SAR Transfer in SAR Target Classification
Hao Zheng, Meiguang Zheng, Zhigang Hu, Liu Yang, Aikun Xu, Tingxuan Chen, Rongchang Zhao, Boyu Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| expand training set with HiFC-GAN pseudo-SAR images | VGG16 · OpenSARShip | accuracy Acc(3): 72.84 -> 78.38 | Improved |
| expand training set with HiFC-GAN pseudo-SAR images | VGG16 · FUSAR-Ship | accuracy Acc(7): 83.22 -> 86.78 | Improved |
| expand training set with HiFC-GAN pseudo-SAR images | AlexNet · OpenSARShip | accuracy Acc(3): 69.97 -> 72.35 | Improved |
| replace CycleGAN-style cycle consistency with hierarchical feature constraints | generator · OpenSARShip | FID: 112.95 -> 69.98 | Improved |
| use CycleGAN pseudo-SAR expansion instead of nothing | VGG16 · OpenSARShip | accuracy: 53.41 -> 67.93 | Mixed |
| increase alpha (more weight on shallow LTC) from 0.4 to 0.8 (ablation) | HiFC-GAN · OpenSARShip | accuracy: 78.38% -> 72.51% | Hurt |
| increase alpha to 0.5 on high-resolution dataset (ablation) | HiFC-GAN · FUSAR-Ship | accuracy: 84.29% -> 86.78% | Improved |
| use 6 instead of 8 constraint feature maps (ablation) | HiFC-GAN · FUSAR-Ship | accuracy: 86.10% -> 86.78% | Improved |
| use LTC and SFM together rather than either alone (ablation) | HiFC-GAN · FUSAR-Ship | FID | Improved |
| add feature constraint loss and identity loss (ablation) | HiFC-GAN · OpenSARShip | image quality | Improved |
Closest papers by meaning
- Ph-GAN: Physics-Inspired GAN for Generating SAR Images Under Limited Data (ICCV 2025 · similarity 0.88)
- SAR-DisentDM: A Semantic-Disentangled Diffusion Model for Limited-Data SAR Image Synthesis (AAAI 2026 · similarity 0.82)
- RIFLE: Backpropagation in Depth for Deep Transfer Learning through Re-Initializing the Fully-connected LayEr (ICML 2020 · similarity 0.81)
- FUSAR-GPT: A Spatiotemporal Feature-Embedded and Two-Stage Decoupled Visual Language Model for SAR Imagery (CVPR 2026 · similarity 0.79)
- Exploring Semantic Feature Discrimination for Perceptual Image Super-Resolution and Opinion-Unaware No-Reference Image Quality Assessment (CVPR 2025 · similarity 0.78)
- Rethinking Infrared Small Target Detection: A Foundation Driven Efficient Paradigm (NeurIPS 2026 · similarity 0.78)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
