Multi-structure model selection via kernel optimisation

Chin, Tat-Jun; Suter, David; Wang, Hanzi

doi:10.1109/cvpr.2010.5539931

Cited by 20 publications

(20 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Other techniques follow a preference-based approach, namely they concentrate on the segmentation side of the problem, from which model estimation follows. Solutions of this type include Residual Histogram Analysis (RHA) [61], J-Linkage [44], Kernel Optimization [5], Tlinkage [27], Random Cluster Model (RCM) [34] and Robust Preference Analysis (RPA) [28]. The problem of fitting multiple models can also be expressed in terms of energy minimization [8,9], as done by PEARL (Propose Expand and Re-estimate Labels) [15] and Multi-X [2].…”

Section: Related Workmentioning

confidence: 99%

Robust Motion Segmentation From Pairwise Matches

Arrigoni

Pajdla

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

In this paper we address a classification problem that has not been considered before, namely motion segmentation given pairwise matches only. Our contribution to this unexplored task is a novel formulation of motion segmentation as a two-step process. First, motion segmentation is performed on image pairs independently. Secondly, we combine independent pairwise segmentation results in a robust way into the final globally consistent segmentation. Our approach is inspired by the success of averaging methods. We demonstrate in simulated as well as in real experiments that our method is very effective in reducing the errors in the pairwise motion segmentation and can cope with large number of mismatches.

show abstract

Section: Related Workmentioning

confidence: 99%

Robust Motion Segmentation From Pairwise Matches

Arrigoni

Pajdla

2019

2019 IEEE/CVF International Conference on Computer Vision (ICCV)

View full text Add to dashboard Cite

show abstract

“…With respect to similar competing methods like [30,31,34], J-linkage has the drawback of requiring the inlier threshols, as RANSAC, and some additional knowledge or processing is needed to determine the number of models. However, it must be noted that:…”

Section: J-linkage Clusteringmentioning

confidence: 99%

“…In other words, J-linkage does not need this piece of information to produce its results, which can then be refined with an educated guess or with an automatic model selection procedure, as in [31].…”

Section: J-linkage Clusteringmentioning

confidence: 99%

“…This allows remove outliers, then the clean data is overclustered with kernel-PCA and spectral clustering and the resulting structures are merged with a sequential ad-hoc scheme that incorporates model selection criteria. In [31] this last stage have been refined with kernel optimization. Residual information is also exploited in [32], a singlemodel estimation technique based on random sampling, where the inlier threshold is not required.…”

Section: J-linkagementioning

confidence: 99%

See 1 more Smart Citation

Image-consistent patches from unstructured points with J-linkage

Toldo

Fusiello

2013

Image and Vision Computing

View full text Add to dashboard Cite

Going from unstructured cloud of points to surfaces is a challenging problem. However, as points are produced by a structure-and-motion pipeline, image-consistency is a powerful clue that comes to the rescue. In this paper we present a method for extracting planar patches from an unstructured cloud of points, based on the detection of image-consistent planar patches with J-linkage, a robust algorithm for multiple models fitting. The method integrates several constraints inside J-linkage, optimizes the position of the points with regard to image-consistency and deploys a hierarchical processing scheme that decreases the computational load. With respect to previous work this approach has the advantage of starting from sparse data. Several results show the e↵ectiveness of the proposed approach.

show abstract

“…This strategy usually has a process of feature pairs similarity analysis in the sampling stage, which are mainly based on feature matching scores or location residuals. Typical methods including dynamic and hierarchical algorithm [28], Top-k [29], Global optimization [30], Multi-GS [17], kernal-based method [31], and Random Cluster Model Sampling (RCMSA) [32,33], etc. Specially, Multi-GS accelerates hypothesis sampling by guiding it with information derived from residual sorting.…”

Section: Related Workmentioning

confidence: 99%

A Robust Transform Estimator Based on Residual Analysis and Its Application on UAV Aerial Images

Cai

Leng

et al. 2018

Remote Sensing

View full text Add to dashboard Cite

Abstract:Estimating the transformation between two images from the same scene is a fundamental step for image registration, image stitching and 3D reconstruction. State-of-the-art methods are mainly based on sorted residual for generating hypotheses. This scheme has acquired encouraging results in many remote sensing applications. Unfortunately, mainstream residual based methods may fail in estimating the transform between Unmanned Aerial Vehicle (UAV) low altitude remote sensing images, due to the fact that UAV images always have repetitive patterns and severe viewpoint changes, which produce lower inlier rate and higher pseudo outlier rate than other tasks. We performed extensive experiments and found the main reason is that these methods compute feature pair similarity within a fixed window, making them sensitive to the size of residual window. To solve this problem, three schemes that based on the distribution of residuals are proposed, which are called Relational Window (RW), Sliding Window (SW), Reverse Residual Order (RRO), respectively. Specially, RW employs a relaxation residual window size to evaluate the highest similarity within a relaxation model length. SW fixes the number of overlap models while varying the length of window size. RRO takes the permutation of residual values into consideration to measure similarity, not only including the number of overlap structures, but also giving penalty to reverse number within the overlap structures. Experimental results conducted on our own built UAV high resolution remote sensing images show that the proposed three strategies all outperform traditional methods in the presence of severe perspective distortion due to viewpoint change.

show abstract

Multi-structure model selection via kernel optimisation

Cited by 20 publications

References 17 publications

Robust Motion Segmentation From Pairwise Matches

Robust Motion Segmentation From Pairwise Matches

Image-consistent patches from unstructured points with J-linkage

A Robust Transform Estimator Based on Residual Analysis and Its Application on UAV Aerial Images

Contact Info

Product

Resources

About