Research map / 3D vision / NeRF and radiance fields
Nef-Net v2: Adapting Electrocardio Panorama in the wild
Zehui Zhan, Yaojun Hu, Jiajing Zhang, Wanchen Lian, Wanqing Wu, Jintai Chen
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 Geometric View Transformer replacing feature averaging encoder-decoder (ablation) | Nef-Net+ · CPSC2018 (3,8,1) synthesis | PSNR: 26.79 -> 29.54 | Improved |
| add query-guided View Encoder (FiLM-style) (ablation) | Nef-Net+ with GeoVT · CPSC2018 (3,8,1) synthesis | PSNR: 29.54 -> 31.19 | Improved |
| add noise perturbation epsilon during training (ablation) | Nef-Net+ · CPSC2018 (3,8,1) synthesis | PSNR: 30.04 -> 31.19 | Improved |
| add Device Calibration stage after pretraining (ablation) | Nef-Net+ · ChinaDB/CPSC2018/Tianchi/PTBXL synthesis | PSNR: gains of 0-1.07 dB (PTBXL 31.15 -> 31.15) | Improved |
| add On-the-fly Calibration stage (ablation) | Nef-Net+ · ChinaDB/CPSC2018/Tianchi/PTBXL synthesis | PSNR: 1.75-2.74 PSNR; CPSC2018 32.08 -> 33.62 | Improved |
| add On-the-fly Calibration on Tianchi SSIM (ablation) | Nef-Net+ · Tianchi synthesis | SSIM: 0.977 -> 0.976 | No clear effect |
| pretrain with Any-Pairs Pretraining on 1% of CPSC2018 data (ablation) | Nef-Net+ · CPSC2018, 1% data (44) | PSNR: 24.68 -> 31.75 | Improved |
| pretrain with Any-Pairs Pretraining on 100% of CPSC2018 data (ablation) | Nef-Net+ · CPSC2018, 100% data (4468) | PSNR: 31.67 -> 32.06 | No clear effect |
| add On-the-fly Calibration under 30 degree lead deviation (ablation) | Nef-Net+ · CPSC2018 with angular deviation | PSNR: 26.53 -> 33.09 | Improved |
| increase number of supervised leads (3 to 12) (ablation) | Nef-Net (synthesis) · Panobench synthesis, inputs I, II, view-18 | PSNR: 21.13 -> 32.98 | Improved |
| increase number of supervised leads (3 to 12) for Nef-Net+ (ablation) | Nef-Net+ (synthesis) · Panobench synthesis, inputs I, II, view-18 | PSNR: 31.63 -> 35.57 | Improved |
| increase number of input leads (3 to 7) (ablation) | Nef-Net and Nef-Net+ · Panobench | PSNR | No clear effect |
Closest papers by meaning
- Extending Pretrained 10-Second ECG Foundation Models to Longer Horizons (NeurIPS 2026 · similarity 0.84)
- Leveraging an ECG Beat Diffusion Model for Morphological Reconstruction from Indirect Signals (NeurIPS 2024 · similarity 0.83)
- GEM: Empowering MLLM for Grounded ECG Understanding with Time Series and Images (NeurIPS 2025 · similarity 0.83)
- Region-Disentangled Diffusion Model for High-Fidelity PPG-to-ECG Translation (AAAI 2024 · similarity 0.82)
- From Token to Rhythm: A Multi-Scale Approach for ECG-Language Pretraining (ICML 2025 · similarity 0.82)
- ECG-Reasoning-Benchmark: A Benchmark for Evaluating Clinical Reasoning Capabilities in ECG Interpretation (NeurIPS 2026 · similarity 0.82)
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.
