rPPG-VQA filters in-the-wild videos using signal-level SNR consensus and scene-level MLLM interference detection, then applies two-stage adaptive sampling to produce unsupervised rPPG models with substantially higher benchmark accuracy.
completely blind
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.CV 2years
2026 2verdicts
UNVERDICTED 2representative citing papers
Q-DeepSight proposes a think-with-image multimodal CoT framework trained via RL with perceptual curriculum rewards and evidence gradient filtering to achieve SOTA IQA performance and enable training-free perceptual refinement in image generation.
citing papers explorer
-
rPPG-VQA: A Video Quality Assessment Framework for Unsupervised rPPG Training
rPPG-VQA filters in-the-wild videos using signal-level SNR consensus and scene-level MLLM interference detection, then applies two-stage adaptive sampling to produce unsupervised rPPG models with substantially higher benchmark accuracy.
-
Q-DeepSight: Incentivizing Thinking with Images for Image Quality Assessment and Refinement
Q-DeepSight proposes a think-with-image multimodal CoT framework trained via RL with perceptual curriculum rewards and evidence gradient filtering to achieve SOTA IQA performance and enable training-free perceptual refinement in image generation.