AI Paper Atlas

Research map / Probabilistic and scientific modeling / Variational and generative sampling

VOLDOR: Visual Odometry From Log-Logistic Dense Optical Flow Residuals

Zhixiang Min, Yiding Yang, Enrique Dunn

CVPR 2020 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Likelihood-free and graphical model estimation · 46 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 VOLDOR instead of MLM-SFMMonocular VO · KITTI odometry training sequences 0–10translation error (%): 2.03 -> 1.32Improved
Use VOLDOR instead of MLM-SFMMonocular VO · KITTI odometry training sequences 0–10rotation error (deg/m): 0.0045 -> 0.0042Improved
Use VOLDOR instead of the strongest listed monocular translation baselineMonocular VO · KITTI odometry test sequences 11–21translation error (%): 1.19 -> 1.65Hurt
Use VOLDOR instead of BVOMonocular VO · KITTI odometry test sequences 11–21rotation error (deg/m): 0.0036 -> 0.0050Hurt
Use VOLDOR instead of the strongest listed baseline on each sequenceMonocular VO · TUM RGB-Dtranslation RMSE (m)Mixed
Use Fisk residuals instead of other tested analytic residual models (ablation)VOLDOR · Depth estimation with few reliable observationsdepth EPEImproved
Use Fisk residuals instead of other residual models (ablation)VOLDOR · Depth estimation with many reliable observationsdepth EPEImproved
Use MIE instead of MLE for depth inference (ablation)VOLDOR with Fisk residuals · Depth estimation across the image sequencedepth accuracyImproved
Use MLE instead of MIE for depth inference (ablation)VOLDOR · Depth estimation with very low rigidnessdepth accuracyImproved
Combine PWC-Net flow, Fisk residuals, and MIE (ablation)VOLDOR · Camera pose estimationpose accuracyImproved
Filter pixels to W^j>5 instead of including all pixels with W^j>0 (ablation)VOLDOR depth estimates · KITTI stereo benchmarkdisparity EPE (px): 1.2304 -> 0.5616Improved
Filter pixels to W^j>5 instead of including all pixels with W^j>0 (ablation)VOLDOR depth estimates · KITTI stereo benchmarkdensity: 100.00% -> 27.07%Hurt

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