AI Paper Atlas

Research map / Label-efficient and robust learning / Test-time adaptation

Meta-rPPG: Remote Heart Rate Estimation Using a Transductive Meta-Learner

Eugene Lee, Evan Chen, Chen-Yi Lee

ECCV 2020 · Label-efficient and robust learning › Test-time adaptation · approach: Extrapolation and single-sample adaptation · 205 citations (OpenAlex)

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 changedWhereMeasuredEffect
add meta-training (inductive inference) (ablation)end-to-end baseline · MAHNOB-HCIMAE (bpm): 7.47 -> 7.42No clear effect
add prototypical distance minimizer at test time (ablation)Meta-rPPG · MAHNOB-HCIMAE (bpm): 7.42 -> 6.05Improved
add synthetic gradient generator at test time (ablation)Meta-rPPG · MAHNOB-HCIMAE (bpm): 7.42 -> 3.88Improved
combine proto and synth transductive gradients (ablation)Meta-rPPG (synth only) · MAHNOB-HCIMAE (bpm): 3.88 -> 3.01Improved
add prototypical distance minimizer at test time (ablation)Meta-rPPG · UBFC-rPPGMAE (bpm): 13.23 -> 7.82Improved
add synthetic gradient generator at test time (ablation)Meta-rPPG · UBFC-rPPGMAE (bpm): 13.23 -> 9.11Improved
combine proto and synth transductive gradients (ablation)Meta-rPPG (proto only) · UBFC-rPPGMAE (bpm): 7.82 -> 5.97Improved
add meta-training without adaptation (ablation)end-to-end baseline · UBFC-rPPGMAE (bpm): 12.78 -> 13.23Hurt
adapt only feature extractor rather than jointly with rPPG estimator (ablation)Meta-rPPG · MAHNOB-HCIMAE (bpm): 4.25 -> 3.01Improved
use 10 adaptation steps instead of 30 at test time (ablation)Meta-rPPG (proto+synth) · MAHNOB-HCIMAE (bpm): 6.20 -> 3.01Improved
increase adaptation steps from 5 to 10 (ablation)Meta-rPPG (proto+synth) · MAHNOB-HCIMAE (bpm): 3.89 -> 3.01Improved
increase adaptation steps for proto-only variant from 10 to 30 (ablation)Meta-rPPG (proto only) · MAHNOB-HCIMAE (bpm): 6.05 -> 6.02No clear effect

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy