AI Paper Atlas

Research map / Detection and segmentation / CNN features

Pixelated Semantic Colorization

Jiaojiao Zhao, Jungong Han, Ling Shao, Cees G. M. Snoek

IJCV 2019 · Detection and segmentation › CNN features · approach: Joint detection and instance segmentation · 92 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 pixelated semantic embedding and generator guidancepixelated colorization PixelCNN · Pascal VOC2012 valsegmentation mean-IoU of colorized images: 67.6 -> 70.3 vs strongest-listed baselines; ours 70.3 vs 68.8Improved
add semantic segmentation branch guiding the color embedding (ablation)pixelated colorization · Pascal VOC2012 and COCO-stuffvisual qualityImproved
add semantic labels to condition the generator (concatenation) (ablation)pixelated semantic colorization · Pascal VOC2012 and COCO-stuffvisual qualityImproved
use channel concatenation instead of feature transformation to inject segmentation into the generator (ablation)pixelated semantic generator · Pascal VOC2012visual qualityImproved
colorize gray images before segmentationDeeplab-ResNet101 · Pascal VOC2012 valmean-IoU: 66.9 -> 70.3Improved
use original color instead of our colorized imageDeeplab-ResNet101 · Pascal VOC2012 valmean-IoU: 70.3 -> 72.1Improved
initialize segmentation from pre-trained colorization instead of scratch (ablation)segmentation branch · Pascal VOC2012validation loss: 0.060 -> 0.043Improved
jointly train segmentation and colorization instead of pre-trained colorization only (ablation)segmentation branch · Pascal VOC2012validation loss: 0.043 -> 0.030Improved
compare PSNR of our semantic method with best baselinepixelated semantic colorization · Pascal VOC2012 valPSNR: 24.56 -> 23.15Hurt
add pixelated semantics to the model (ablation)pixelated colorization · Pascal VOC2012 valsample diversity (MS-SSIM)No clear effect
use ours versus best diverse baseline for naturalnesspixelated semantic colorization · VOC + COCO-stuffnaturalness (%): 89.89 -> 94.65Improved

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