AI Paper Atlas

Research map / Image generation / Autoregressive and latent generation

Latent Semantic Consensus for Deterministic Geometric Model Fitting

Guobao Xiao, Jun Yu, Jiayi Ma, Deng-Ping Fan, Ling Shao

TPAMI 2024 · Image generation › Autoregressive and latent generation · approach: Latent diffusion with learned tokenizers · 15 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
use LSC-SA latent-semantic sampling instead of RANSAC samplingMSSC-style model selection · Hopkins155mean segmentation error (%): 0.83 -> 0.71Improved
sample from affine, homography and fundamental models (Subset) instead of homography onlyLSC · Hopkins155mean segmentation error (%): 0.71 -> 0.31Improved
use LSC instead of Subset for speedSubset · Hopkins155CPU time (s): 2621.80 -> 1508.19Improved
use LSC instead of RANSACRANSAC · KITTI sequence 02translation error (m): 4.01 -> 2.24Improved
use LSC instead of RANSACRANSAC · KITTI sequence 00CPU time (s): 1.07 -> 0.33Improved
use LSC (high-quality hypotheses) instead of CLSA selection on poor hypothesesCLSA · 4 circles syntheticsegmentation error (%): 38.55 -> 1.25Improved
use LSC instead of CLSACLSA · 4 lines syntheticsegmentation error (%): 2.28 -> 2.63Hurt
use LSC instead of RansaCovRansaCov · 5 circles syntheticsegmentation error (%): 0.23 -> 0.33Hurt
set threshold psi to 0.05 for homography estimation (ablation)LSC · AdelaideRMF homographysegmentation errorImproved
set threshold psi to 0.01 for fundamental matrix estimation (ablation)LSC · AdelaideRMF fundamental matrixsegmentation errorImproved
set threshold beta to 0.8 (ablation)LSC · AdelaideRMFsegmentation errorImproved
vary psi from 0.01 to 0.10 (ablation)LSC · MS-COCO-F homographysegmentation errorNo clear effect

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