AI Paper Atlas

Research map / 3D vision / Depth and stereo

Distortion-Aware Convolutional Filters for Dense Prediction in Panoramic Images

Keisuke Tateno, Nassir Navab, Federico Tombari

ECCV 2018 · 3D vision › Depth and stereo · approach: Self-supervised monocular depth · 203 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
replace standard convolution with distortion-aware convolution on equirectangular inputFCRN depth prediction · Stanford 2D-3D-S depthrel error: 0.201 -> 0.176Improved
replace cube map rectification with distortion-aware convolutionFCRN depth prediction · Stanford 2D-3D-S depthrel error: 0.220 -> 0.176Improved
distortion-aware convolution versus cube map on rms and log10FCRN depth prediction · Stanford 2D-3D-S depthrms / log10 error: 0.371 -> 0.369 rms; 0.0818 -> 0.0829 log10No clear effect
replace standard convolution with distortion-aware convolution when trained on NYU v2 perspective imagesFCRN depth prediction · Stanford 2D-3D-S depth (trained on NYU v2)rel error: 0.604 -> 0.517Improved
replace cube map with distortion-aware convolution when trained on NYU v2FCRN depth prediction · Stanford 2D-3D-S depth (trained on NYU v2)rms error: 0.692 -> 0.578Improved
replace standard convolution with distortion-aware convolutionFCRN segmentation · Stanford 2D-3D-S semantic segmentationmIoU (%): 32.63 -> 34.56Improved
replace cube map with distortion-aware convolutionFCRN segmentation · Stanford 2D-3D-S semantic segmentationmIoU (%): 33.82 -> 34.56Improved
replace standard convolution with distortion-aware convolution on floor classFCRN segmentation · Stanford 2D-3D-S semantic segmentation (floor class)IoU (%): 78.01 -> 83.40Improved
distortion-aware convolution versus standard convolution on clutter classFCRN segmentation · Stanford 2D-3D-S semantic segmentation (clutter class)IoU (%): 18.84 -> 19.52No clear effect
distortion-aware convolution versus cube map on wall classFCRN segmentation · Stanford 2D-3D-S semantic segmentation (wall class)IoU (%): 61.77 -> 57.17Hurt

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