Research map / Label-efficient and robust learning / Recognition and long-tail
Signal Enhancement via Multi-view Dynamic Representation and Alignment-aware Fusion
Zikun Jin, Yuhua Qian, Xinyan Liang, Jiaqian Zhang, Jinpeng Yuan, Shen Hu, Haijun Geng, Honghong Cheng
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 |
|---|---|---|---|
| add STFrFT view alongside STFT (vs single STFT view) (ablation) | FracFusion · VoiceBank+DEMAND | PESQ, CBAK, COVL, SSNR | Improved |
| add Pearson Channel Fusion (PCF) (ablation) | FracFusion · VoiceBank+DEMAND | PESQ, CSIG | Improved |
| add Efficient Align Fusion (EAF) (ablation) | FracFusion · VoiceBank+DEMAND | SSNR | Improved |
| add FSCAM structure-aware feature extractor (ablation) | FracFusion · VoiceBank+DEMAND | PESQ, CSIG, CBAK, COVL, SSNR, STOI | Improved |
| add structure-aware SA Block (ablation) | FracFusion · VoiceBank+DEMAND | PESQ, CSIG, CBAK, COVL, SSNR, STOI | Improved |
| FracFusion versus best baseline DR-DifuSE on CSIG | FracFusion · VoiceBank+DEMAND | CSIG: 4.38 -> 4.56 | Improved |
| FracFusion versus MFSE (STFT+STFrFT with fixed fusion) on SSNR | FracFusion · VoiceBank+DEMAND | SSNR: 10.63 -> 10.52 | Hurt |
| FracFusion versus best baseline on EARS-WHAM! PESQ | FracFusion · EARS-WHAM! | PESQ: 2.39 -> 2.79 | Improved |
Closest papers by meaning
- Representing Sounds as Neural Amplitude Fields: A Benchmark of Coordinate-MLPs and a Fourier Kolmogorov-Arnold Framework (AAAI 2025 · similarity 0.82)
- MERLIN: Building Low-SNR Robust Multimodal LLMs for Electromagnetic Signals (CVPR 2026 · similarity 0.81)
- Learning High-Frequency Functions Made Easy with Sinusoidal Positional Encoding (ICML 2024 · similarity 0.81)
- Misalignment-Robust Frequency Distribution Loss for Image Transformation (CVPR 2024 · similarity 0.80)
- Vocos: Closing the gap between time-domain and Fourier-based neural vocoders for high-quality audio synthesis (ICLR 2024 · similarity 0.80)
- Audio Super-Resolution with Latent Bridge Models (NeurIPS 2025 · similarity 0.80)
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.
