AI Paper Atlas

Research map / Image generation / Image restoration

Deep Single Image Camera Calibration With Radial Distortion

Manuel Lopez, Roger Mari, Pau Gargallo, Yubin Kuang, Javier Gonzalez-Jimenez, Gloria Haro

CVPR 2019 · Image generation › Image restoration · approach: Image harmonization and color transformation · 99 citations (OpenAlex)

Check 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
Use the learned CNN rather than the plumb-line method for undistortionSingle-image radial distortion correction · Synthetic test set (16,500 images)images with lower undistortion MSE: 89%Improved
Use the learned CNN rather than the plumb-line method for undistortionSingle-image radial distortion correction · Outdoor imageryimages where proposed CNN performs best: more than 90%Improved
Use apparent distortion k̂1 and distorted offset ρ instead of k1 and undistorted offset τ (ablation)Proposed camera-calibration network · Synthetic test setparameter prediction accuracyImproved
Train with bearing loss instead of unit-weight Huber losses (ablation)Proposed camera-calibration network · Validation setk̂1 prediction errorHurt
Train with bearing loss instead of unit-weight Huber losses (ablation)Proposed camera-calibration network · Validation setprediction error for parameters other than k̂1No clear effect
Use unit weights for the Huber loss components rather than the bearing loss (ablation)Proposed camera-calibration network · Validation setparameter prediction errorImproved

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