Activity counts measure attention, not model quality. Missing counts are unknown. Publication status follows the linked source; a listed venue alone does not confirm peer review.
MSTrans is a multi-scale spatio-temporal transformer framework for remote photoplethysmography from face video, described in Computer Vision and Image Understanding (DOI 10.1016/j.cviu.2026.104983). Public abstract text and official code were not reachable when this entry was prepared, so method details and reported errors are omitted here.
- Signals
- RGB
- Form factor
- camera_rppg
- License
- Paper closed access (Elsevier)
- Year
- 2027
- Architecture
multi-scale spatio-temporal transformer for rPPG
- Type
- Task model
- Modalities
- RGB
- Form factor
- Camera rPPG
- Parameters
- Not reported
- Architecture
- multi-scale spatio-temporal transformer for rPPG
- Venue
- Computer Vision and Image Understanding (Elsevier), vol. Jan 2027, DOI 10.1016/j.cviu.2026.104983
- Year
- 2027
- arXiv
- Not reported
- Code
- Not reported
- Weights
- Not reported
- Licence
- Paper closed access (Elsevier)
- Caveats
- Not reported
- Datasets
- Not reported
- Added
- Not reported
- Updated
- Oct 27, 2026

How it works
- Architecture
- multi-scale spatio-temporal transformer for rPPG
- Inputs
- RGB
- Approach
- rppg · model · transformer · multi scale
Datasets
Cite
APA
MSTrans. (2027). Computer Vision and Image Understanding. https://doi.org/10.1016/j.cviu.2026.104983
BibTeX
@article{mstrans2027mstrans,
title = {{MSTrans}},
year = {2027},
journal = {Computer Vision and Image Understanding},
doi = {10.1016/j.cviu.2026.104983}
}Built from the hub record. Check it against the publisher before submitting.
History
Record updated
- RhythmFormerShared topic: transformerperiodic sparse attention Transformer for rPPG2024
- FLOWShared topic: modelfeature-level optimal warping for domain-generalized rPPG2026
- MS-rPPGShared topic: modelmulti-spectral (RGB + NIR) state space model for in-vehicle rPPG2026
- PALC-NetShared topic: modelphysiology-aware lightweight contrastive network for unsupervised rPPGRGB2026
