AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Effects of Image Degradation and Degradation Removal to CNN-Based Image Classification

Yanting Pei, Yaping Huang, Qi Zou, Xingyuan Zhang, Song Wang

TPAMI 2019 · Label-efficient and robust learning › Recognition and long-tail · approach: Loss design and classifier theory · 290 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
train on clear images instead of matched degraded imagesVGGNet-16 · Haze beta=5 testaccuracy (%): 67.8 -> 33.0Hurt
train on clear images instead of matched motion-blurred imagesVGGNet-16 · Motion blur (40,40) testaccuracy (%): 59.0 -> 11.2Hurt
increase motion blur from (0,0) to (40,40) with matched trainingVGGNet-16 · Motion bluraccuracy (%): 81.0 -> 59.0Hurt
use VGGNet-16 instead of AlexNetCNN classifier · Caltech-256 clearaccuracy (%): 71.7 -> 81.0Improved
mix all degradation levels in trainingVGGNet-16 · Haze beta=5 testaccuracy (%): 67.8 -> 69.2Improved
mix degraded levels into training, evaluated on clear imagesVGGNet-16 · Caltech-256 clear test (hazy mix)accuracy (%): 81.0 -> 79.7Hurt
add dehazing pre-processing at test time (train clear)VGGNet-16 · Haze beta=5 testaccuracy (%): 33.0 -> 43.0Improved
add deblurring pre-processing at test time (train clear)VGGNet-16 · Motion blur (15,15) testaccuracy (%): 45.1 -> 50.3Improved
add distortion correction at test time (train clear)VGGNet-16 · Fish-eye e=1.75 testaccuracy (%): 25.4 -> 20.0Hurt
train on real hazy images instead of clear imagesVGGNet-16 · Haze-20 testaccuracy (%): 62.3 -> 81.2Improved
combine hazy and clear images in trainingVGGNet-16 · HazeClear-20 testaccuracy (%): 88.7 -> 97.7Improved

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