AI Paper Atlas

Research map / 3D vision / Human pose estimation

Minimal Solutions to Relative Pose Estimation From Two Views Sharing a Common Direction With Unknown Focal Length

Yaqing Ding, Jian Yang, Jean Ponce, Hui Kong

CVPR 2020 · 3D vision › Human pose estimation · approach: Learned camera pose estimation · 30 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 4-point common-direction solver instead of the 6-point solverRelative pose estimation with shared unknown focal length · First real-data sequencemean rotation error (degrees): 1.0708 -> 0.2224Improved
Use the 4-point common-direction solver instead of the 6-point solverRelative pose estimation with shared unknown focal length · First real-data sequencemean translation error (degrees): 5.9041 -> 1.7173Improved
Use the 4-point common-direction solver instead of the 6-point solverRelative pose estimation with shared unknown focal length · First real-data sequencemean focal length error (%): 41.15 -> 11.31Improved
Use the 4-point common-direction solver instead of the 6-point solverRelative pose estimation with shared unknown focal length · First real-data sequencemean RANSAC iterations: 1423 -> 237Improved
Use the 5-point common-direction solver instead of the 7-point solverRelative pose estimation with different unknown focal lengths · First real-data sequencemean rotation error (degrees): 4.0895 -> 0.5607Improved
Use the 5-point common-direction solver instead of the 7-point solverRelative pose estimation with different unknown focal lengths · First real-data sequencemean translation error (degrees): 13.5088 -> 4.8110Improved
Use the 5-point common-direction solver instead of the 7-point solverRelative pose estimation with different unknown focal lengths · First real-data sequencemean focal length error (%): 43.35 -> 31.37Improved
Use the 5-point common-direction solver instead of the 7-point solverRelative pose estimation with different unknown focal lengths · First real-data sequencemean RANSAC iterations: 2474 (85) -> 452 (180)Improved
Use the 5-point common-direction solver instead of the 7-point solverRelative pose estimation with different unknown focal lengths · First real-data sequencemean inliers: 696 -> 707Improved
Use the 4-point solver rather than the 3.5-point homography solverRelative pose estimation with shared unknown focal length · First real-data sequencemean focal length error (%): 42.02 -> 11.31Improved
Use the 5-point solver rather than the H4pt homography solverRelative pose estimation with different unknown focal lengths · First real-data sequencemean focal length error (%): 57.47 -> 31.37Improved

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