AI Paper Atlas

Research map / Image generation / Image restoration

Single Image Intrinsic Decomposition without a Single Intrinsic Image

Wei-Chiu Ma, Hang Chu, Bolei Zhou, Raquel Urtasun, Antonio Torralba

ECCV 2018 · Image generation › Image restoration · approach: Dehazing and deraining · 69 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 disentangled reconstruction loss Ldis (ablation)unsupervised intrinsic net (Lrec only) · MIT Intrinsicsalbedo LMSE: 0.0158 -> 0.0141Improved
add disentangled reconstruction loss Ldis (ablation)unsupervised intrinsic net (Lrec only) · MIT Intrinsicsalbedo MSE: 0.0362 -> 0.0346Improved
add Retinex loss (ablation)Lrec + Ldis model · MIT Intrinsicsalbedo LMSE: 0.0141 -> 0.0116Improved
add Retinex loss (ablation)Lrec + Ldis model · MIT Intrinsicsshading LMSE: 0.0098 -> 0.0095Improved
add Retinex loss (ablation)Lrec + Ldis model · MIT Intrinsicsalbedo MSE: 0.0346 -> 0.0313Improved
fine-tune unsupervised model with ground truth (100%)Our method · ShapeNetaverage LMSE: 0.0060 -> 0.0020Improved
fine-tune unsupervised model with ground truth (100%)Our method · MIT Intrinsicsaverage LMSE: 0.0105 -> 0.0063Improved
fine-tune unsupervised model with ground truth (100%)Our method · MIT Intrinsicsaverage MSE: 0.0260 -> 0.0130Improved
fine-tune unsupervised model with ground truthOur method · MIT Intrinsics (illumination invariance)MPRE (x10^-4): 3.2341 -> 2.4151Improved
use learned soft assignment mask instead of traditional Retinex for gradient classificationOur method · MIT Intrinsicsgradient pseudo-GT error: 2.6 x 10-4 -> 1.7 x 10-4Improved
use only 50% of ground-truth supervision during fine-tuningOur method · ShapeNetMSE/LMSE: state-of-the-art with only 50% of ground truthImproved

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