AI Paper Atlas

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

Generalized Differentiable RANSAC

Tong Wei, Yash Patel, Alexander Shekhovtsov, Jiri Matas, Daniel Barath

ICCV 2023 · Probabilistic and scientific modeling › Variational and generative sampling · approach: Gradient estimation for discrete variables · 51 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
train sampling weights end-to-end with differentiable RANSAC instead of using the pretrained CLNet, OANet or NG-RANSAC∇-RANSAC · PhotoTourism F estimation (authors' pretrained baselines)F1 score (%): 45.80 -> 48.12Improved
train with normalized 8PC solver instead of 7PC (ablation)∇-RANSAC (F estimation) · PhotoTourism F estimationF1 score (%): 46.04 -> 47.21Improved
train end-to-end instead of KL initialization only (ablation)∇-RANSAC (F estimation) · PhotoTourism F estimationmedian epipolar error (px): 1.73 -> 1.00Improved
use 5PC solver for E training rather than 8PC (ablation)∇-RANSAC (E estimation) · PhotoTourism E estimationAUCImproved
use epipolar error loss instead of pose error loss for E estimation (ablation)∇-RANSAC · PhotoTourism E estimationAUCImproved
use learned GS sampler weights instead of uniform sampling or SNN-guided GS (ablation)∇-RANSAC with MAGSAC++ · PhotoTourism E estimationAUCImproved
replace 8PC with 7PC for F training (ablation)∇-RANSAC · PhotoTourism F estimationF1 score (%)No clear effect
replace Gumbel-softmax objective with DSAC soft model selection (ablation)∇-RANSAC · PhotoTourism F estimationF1 score (%): 48.12 -> 40.95Hurt
replace objective with SoftAM (ablation)∇-RANSAC · PhotoTourism F estimationF1 score and run-time: 48.12 -> 48.17 F1; 28.66 -> 47.46 msNo clear effect
use NG-RANSAC gradient approximation instead of GS relaxation (ablation)∇-RANSAC · PhotoTourism F estimationF1 score (%): 48.12 -> 41.26Hurt
fine-tune LoFTR through ∇-RANSACLoFTR with MAGSAC-PROSAC · PhotoTourism F estimationavg. epipolar error (px): 13.06 -> 10.31Improved
fine-tune LoFTR through ∇-RANSACLoFTR with OpenCV-RANSAC · PhotoTourism F estimationF1 score (%): 64.07 -> 64.43No 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