Matching with PROSAC - progressive sample consensus

By Ondřej Chum, Jiri Matas. Computer Vision and Pattern Recognition, 2005

1,354 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/e4f4c2bad99161cd10b063dec855902d2da0ee72

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Performance Evaluation of RANSAC Family (2009, background): “PROSAC (Progressive SAC) [6] uses prior knowledge more deeply.”
Image Matching Across Wide Baselines: From Paper to Practice (2020, methodology): “Improvements to this classical framework include local optimization [28], using likelihood instead of reprojection (MLESAC) [103], speed-ups using probabilistic sampling of hypotheses (PROSAC) [27], degeneracy check using homographies (DEGENSAC) [30], Graph Cut as a local optimization (GC-RANSAC) [1”
A Comparative Analysis of RANSAC Techniques Leading to Adaptive Real-Time Random Sample Consensus (2008, methodology): “First, we provide a comparative analysis of the important RANSAC algorithms that have been proposed over the past few years [2,3,4,5,7,8,11].”
Robust Feature Matching for Remote Sensing Image Registration via Locally Linear Transforming (2015, methodology): “the classical RANSAC algorithm [39] and analogous robust hypothesize-and-verify methods [40], [41] that typically rely on parametric models, and vector field consensus [34], [42], robust point matching via L2E [43], graph shift (GS) [44], and identifying correspondence function (ICF) [45], which are”
A review of multimodal image matching: Methods and applications (2021, methodology): “Many improvements based on RANSAC mainly focus on verifying the model quality, such as using a maximum likelihood procedure in MLESAC [307,308] or modifying the sampling strategy as inspired from the specific properties of inliers, such as spatially consistency-based [312], by using a grouping strat”
Worldwide Pose Estimation Using 3D Point Clouds (2012, neutral): “match based on cues such as proximity of matched features; PROSAC [24], which samples based on a matching quality measure; and GroupSAC [25], which selects samples using cues such as image segmentation.”
Robust Point Matching via Vector Field Consensus (2014, background): “A uthor M anuscript A uthor M anuscript Index Terms Point correspondence; outlier removal; matching; regularization I. Introduction Establishing reliable correspondence between two images is a fundamental problem in computer vision and it is a critical prerequisite in a wide range of applications…”
Learning to Find Good Correspondences (2017, neutral): “PROSAC [7] speeds up the estimation process.”
Image Matching from Handcrafted to Deep Features: A Survey (2020, background): “PROSAC (Chum andMatas 2005) exploits a priori predicted inlier probability, and EVSAC (Fragoso et al. 2013) uses an estimate of confidence with extreme value theory of the correspondences.”
DISK: Learning local features with policy gradient (2020, methodology): “To pick the best checkpoint, we evaluate performance in terms of pose estimation accuracy in stereo, with DEGENSAC [8].”
Applying Deep Learning in Augmented Reality Tracking (2016, methodology): “Usualy robust methods such as RANSAC [15] or one of its variants like PROSAC [16] are employed.”
The structure-from-motion reconstruction pipeline - a survey with focus on short image sequences (2010, positive): “The random sample consensus [13] has received many improvements such as MSAC and MLESAC [43], Locally Optimized RANSAC (LO-RANSAC) [8, 9], PROgressive ranSAC (PROSAC) [7], and more recently QDEGSAC [16] to detect quasi degenerate cases.”
DSAC — Differentiable RANSAC for Camera Localization (2016, methodology): “Dozens of variations of RANSAC exist [39, 8, 28, 7].”
SIFT Meets CNN: A Decade Survey of Instance Retrieval (2016, methodology): “on. The GIST feature [66] shares a similar idea with SPM. These methods lack invariance to image transformations. 3.3.1 SIFT-based In image retrieval, global spatial veriﬁcation based on RANSAC [11], [78] is effective in re-ranking a subset of top-ranked images, but is prone to efﬁciency problems. A”
A comparative analysis of SIFT, SURF, KAZE, AKAZE, ORB, and BRISK (2018, methodology): “Random Sample Consensus (RANSAC) [10], M-estimator Sample Consensus (MSAC) [11], and Progressive Sample Consensus (PROSAC) [12] are some of the robust probabilistic methods used for removing the outliers from matched features and fitting the transformation function (in terms of homography matrix).”
MAGSAC: Marginalizing Sample Consensus (2018, methodology): “NAPSAC assumes that the inliers are spatially coherent, PROSAC exploits an a priori predicted inlier probability of the points and EVSAC estimates a confidence in each of them.”
Real-Time Detection and Tracking for Augmented Reality on Mobile Phones (2010, methodology): “The remaining matches are used in a PROSAC scheme [4] to estimate a homography between the model points of the planar target and the input image.”
Visual Place Recognition with Repetitive Structures (2015, neutral): “dard RANSAC [58] and PROSAC [59] but observed no significant difference in performance.”
Graph-Cut RANSAC (2017, methodology): “PROSAC [4], then computes the parameters θk of the implied model and its support”