Research map / Video / Video segmentation and tracking
PHASE-Net: Physics-Grounded Harmonic Attention System for Efficient Remote Photoplethysmography Measurement
Bo Zhao, Dan Guo, Junzhe Cao, Yong Xu, Bochao Zou, Tao Tan, Yue Sun, Zitong Yu
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 |
|---|---|---|---|
| use PHASE-Net physics-grounded causal TCN model instead of prior best on UBFC-rPPG | rPPG estimation · UBFC-rPPG (intra-dataset) | MAE (bpm): 0.16 -> 0.15 | Improved |
| use PHASE-Net instead of RhythmFormer | rPPG estimation · PURE (intra-dataset) | MAE (bpm): 0.27 -> 0.14 | Improved |
| use PHASE-Net in leave-one-out transfer to PURE | rPPG estimation · Leave-one-out, others to PURE | MAE (bpm): 21.11 -> 2.86 | Improved |
| use PHASE-Net on MMPD intra-dataset versus RhythmFormer | rPPG estimation · MMPD (intra-dataset) | MAE (bpm): 4.69 -> 4.78 | No clear effect |
| add GTCN (gated TCN) (ablation) | PHASE-Net · UBFC-rPPG | RMSE (bpm): 1.26 -> 0.90 | Improved |
| add adaptive spatial attention (ablation) | PHASE-Net · PURE | RMSE (bpm): 0.36 -> 0.14 | Improved |
| add ZAS module (ablation) | PHASE-Net · UBFC-rPPG | RMSE (bpm): increase of about 0.14 bpm when removed | Improved |
| increase TCN layers from 1 to 3 (ablation) | PHASE-Net · UBFC-rPPG | MAE (bpm): 0.51 -> 0.15 | Improved |
| add a fourth TCN layer (ablation) | PHASE-Net · UBFC-rPPG | MAE/RMSE (bpm): 0.15/0.53 -> 0.25/0.69 | Hurt |
| set auxiliary loss weight lambda to 0.1 instead of 0.01 (ablation) | PHASE-Net · UBFC-rPPG | MAE/RMSE (bpm): 0.32/0.93 -> 0.15/0.53 | Improved |
| increase lambda from 0.1 to 1 (ablation) | PHASE-Net · UBFC-rPPG | MAE/RMSE (bpm): 0.15/0.53 -> 0.43/1.27 | Hurt |
| ZAS block size 2 and 2 swapped channel groups versus other values (ablation) | PHASE-Net ZAS · UBFC, PURE, BUAA, MMPD | RMSE (bpm) | Improved |
Later papers that compared with this method
1 later accepted paper reports PHASE-Net as a baseline in their results tables; 0 of them report a better number on most of the results they share with it.
- FreqPhys: Repurposing Implicit Physiological Frequency Prior for Robust Remote Photoplethysmography (ECCV 2026 · better on 2 of 4 shared results)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- PhysFormer — PhysFormer: Facial Video-Based Physiological Measurement With Temporal Difference Transformer (CVPR 2022) · better on 5 of 5 shared results
- Contrast-Phys+ — Contrast-Phys: Unsupervised Video-Based Remote Physiological Measurement via Spatiotemporal Contrast (ECCV 2022) · better on 2 of 2 shared results
- PhysDiff — PhysDiff: Physics-Guided Human Motion Diffusion Model (ICCV 2023) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- PhysFormer: Facial Video-Based Physiological Measurement With Temporal Difference Transformer (CVPR 2022 · similarity 0.91)
- RhythmMamba: Fast, Lightweight, and Accurate Remote Physiological Measurement (AAAI 2025 · similarity 0.91)
- PhysDiff: Physiology-based Dynamicity Disentangled Diffusion Model for Remote Physiological Measurement (AAAI 2025 · similarity 0.89)
- Remote Photoplethysmography in Real-World and Extreme Lighting Scenarios (CVPR 2025 · similarity 0.89)
- FactorizePhys: Matrix Factorization for Multidimensional Attention in Remote Physiological Sensing (NeurIPS 2024 · similarity 0.89)
- FacePhys: State of the Heart Learning (NeurIPS 2026 · similarity 0.87)
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.
