AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

Transformation Consistency Regularization – A Semi-Supervised Paradigm for Image-to-Image Translation

Aamir Mustafa, Rafal K. Mantiuk

ECCV 2020 · Label-efficient and robust learning › Semi-supervised learning · approach: Pseudo-labeling for specialized tasks · 33 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 TCR semi-supervised training on unlabeled data (10% labeled)Colorization network · Image colorization (Places)PSNR: +3.1 dBImproved
add TCR semi-supervised training on unlabeled data (10% labeled)DnCNN · Image denoisingPSNR: +3.0 dBImproved
add TCR semi-supervised training with 1% labeled framesColorization network · Big Buck Bunny movie colorizationPSNR: +6.1 dBImproved
add TCR semi-supervised training with 1% labeled framesEDSR · Elephants Dream movie super-resolutionPSNR: +5.7 dB (4x), +2.9 dB (3x)Improved
add TCR loss on labeled data only (ablation)Supervised baselines · Colorization, denoising, super-resolutionreconstruction quality (PSNR/FSIM)Improved
replace geometric data augmentation with TCR (labeled only) (ablation)Supervised baselines · Colorization, denoising, super-resolutionreconstruction quality (PSNR/FSIM)Improved
add TCR so denoised images remap to the clean manifoldDnCNN (10% labeled) · Manifold validation, denoisingfraction of outputs classified as clean: 44.3% -> 97.9%Improved
increase unlabeled-to-labeled ratio r per batch up to 10 (ablation)TCR · Colorization and denoising (10% and 20% labeled)PSNRImproved
set regularization weight lambda to 10 (searched 10^-1 to 10^4) (ablation)TCR · Colorization and denoising (10% and 20% labeled)PSNRImproved
add TCR with perceptual loss instead of L2 (ablation)Super-resolution model · Super-resolution (perceptual loss)PSNRImproved
add TCR with L1 loss instead of L2 (ablation)Super-resolution model · Super-resolution (L1 loss)PSNRImproved
use perceptual loss instead of pixel L2 loss (ablation)Super-resolution model · Super-resolutionPSNRHurt

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