DGSAC: Density Guided Sampling and Consensus

Tiwari, Lokender; Anand, Sandeep

doi:10.1109/wacv.2018.00112

Cited by 13 publications

(10 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…An ensemble of these layer-wise generative classifiers is used to make the final prediction by performing a Borda count-based rank-aggregation. Ranking preferences have been used extensively in robust fitting problems in computer vision (Chin et al, 2011;jun Chin et al, 2009;Tiwari & Anand, 2018), and we show its effectiveness in introducing robustness in DNNs against adversarial attacks.…”

Section: Regroup Methodologysupporting

confidence: 53%

REGroup: Rank-aggregating Ensemble of Generative Classifiers for Robust Predictions

Tiwari¹,

Madan²,

Anand³

et al. 2022

2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

Self Cite

View full text Add to dashboard Cite

Deep Neural Networks (DNNs) are often criticized for being susceptible to adversarial attacks. Most successful defense strategies adopt adversarial training or random input transformations that typically require retraining or fine-tuning the model to achieve reasonable performance. In this work, our investigations of intermediate representations of a pre-trained DNN lead to an interesting discovery pointing to intrinsic robustness to adversarial attacks. We find that we can learn a generative classifier by statistically characterizing the neural response of an intermediate layer to clean training samples. The predictions of multiple such intermediate-layer based classifiers, when aggregated, show unexpected robustness to adversarial attacks. Specifically, we devise an ensemble of these generative classifiers that rank-aggregates their predictions via a Borda count-based consensus. Our proposed approach uses a subset of the clean training data and a pre-trained model, and yet is agnostic to network architectures or the adversarial attack generation method. We show extensive experiments to establish that our defense strategy achieves state-of-the-art performance on the ImageNet validation set.

show abstract

Section: Regroup Methodologysupporting

confidence: 53%

REGroup: Rank-aggregating Ensemble of Generative Classifiers for Robust Predictions

Tiwari¹,

Madan²,

Anand³

et al. 2022

2022 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV)

Self Cite

View full text Add to dashboard Cite

show abstract

“…In this section, we firstly evaluate the influence of the different components of HRMP on the performance. Then, we compare the proposed HRMP with eleven state-of-theart model fitting methods, including J-Linkage [15], KF [8], AKSWH [9], T-Linkage [16], RPA [17], DPA [20], RansaCov [18], TSMP [29], HVF [30], DGSAC [19] and MSHF [10] on both challenging synthetic data and real images. For fair comparison, the model hypotheses are generated by employing the proximity sampling [15] for all the fitting methods.…”

Section: Methodsmentioning

confidence: 99%

“…Generally speaking, robust geometric model fitting methods [8]- [10], [15]- [18] can be coarsely classified into the consensus analysis based methods and the preference analysis based methods. The consensus analysis based fitting methods (e.g., adaptive kernel-scale weighted hypotheses (AKSWH) [9], density guided sampling and consensus (DGSAC) [19], random sample coverage (RansaCov) [18] and mode-seeking on hypergraphs fitting (MSHF) [10]) exploit the consensus set corresponding to each model hypothesis for distinguishing inliers from outliers. The preference analysis based fitting methods (e.g., kernel fitting (KF) [8], Jaccard distance based clustering (J-linkage) [15], Tanimoto distance based clustering (T-linkage) [16], robust preference analysis (RPA) [17] and density preference analysis (DPA) [20]) focus on the preference of each data point to the significant model hypotheses for clustering data points.…”

Section: Introductionmentioning

confidence: 99%

Hierarchical Representation Via Message Propagation for Robust Model Fitting

Lin

Wang

Xiao

et al. 2021

IEEE Trans. Ind. Electron.

View full text Add to dashboard Cite

In this paper, we propose a novel hierarchical representation via message propagation (HRMP) method for robust model fitting, which simultaneously takes advantages of both the consensus analysis and the preference analysis to estimate the parameters of multiple model instances from data corrupted by outliers, for robust model fitting. Instead of analyzing the information of each data point or each model hypothesis independently, we formulate the consensus information and the preference information as a hierarchical representation to alleviate the sensitivity to gross outliers. Specifically, we firstly construct a hierarchical representation, which consists of a model hypothesis layer and a data point layer. The model hypothesis layer is used to remove insignificant model hypotheses and the data point layer is used to remove gross outliers. Then, based on the hierarchical representation, we propose an effective hierarchical message propagation (HMP) algorithm and an improved affinity propagation (IAP) algorithm to prune insignificant vertices and cluster the remaining data points, respectively. The proposed HRMP can not only accurately estimate the number and parameters of multiple model instances, but also handle multi-structural data contaminated with a large number of outliers. Experimental results on both synthetic data and real images show that the proposed HRMP significantly outperforms several state-of-the-art model fitting methods in terms of fitting accuracy and speed.

show abstract

“…For example, Barath et al [4] presented the MAGSAC algorithm that does not require a single inlier-outlier threshold such as RANSAC. By exploiting the residual density, Tiwari and Anand [56] introduced the DGSAC algorithm. In the work of Ranftl and Koltun [44] outliers are removed via geometric model estimation and the underlying fundamental matrix is computed using deep neural networks.…”

Section: Related Workmentioning

confidence: 99%

Dense Feature Matching Based on Homographic Decomposition

2022

View full text Add to dashboard Cite

Finding robust and accurate feature matches is a fundamental problem in computer vision. However, incorrect correspondences and suboptimal matching accuracies lead to significant challenges for many real-world applications. In conventional feature matching, corresponding features in an image pair are greedily searched using their descriptor distance. The resulting matching set is then typically used as input for geometric model fitting methods to find an appropriate fundamental matrix and filter out incorrect matches. Unfortunately, this basic approach cannot solve all practical problems, such as fundamental matrix degeneration, matching ambiguities caused by repeated patterns and rejection of initially mismatched features without further reconsideration. In this paper we introduce a novel matching pipeline, which addresses all of the aforementioned challenges at once: First, we perform iterative rematching to give mismatched feature points a further chance for being considered in later processing steps. Thereby, we are searching for inliers that exhibit the same homographic transformation per iteration. The resulting homographic decomposition is used for refining matches, occlusion detection (e.g. due to parallaxes) and extrapolation of additional features in critical image areas. Furthermore, Delaunay triangulation of the matching set is utilized to minimize the repeated pattern problem and to implement focused matching. Doing so, enables us to further increase matching quality by concentrating on local image areas, defined by the triangular mesh. We present and discuss experimental results with multiple real-world matching datasets. Our contributions, besides improving matching recall and precision for image processing applications in general, also relate to use cases in image-based computer graphics.

show abstract

DGSAC: Density Guided Sampling and Consensus

Cited by 13 publications

References 23 publications

REGroup: Rank-aggregating Ensemble of Generative Classifiers for Robust Predictions

REGroup: Rank-aggregating Ensemble of Generative Classifiers for Robust Predictions

Hierarchical Representation Via Message Propagation for Robust Model Fitting

Dense Feature Matching Based on Homographic Decomposition

Contact Info

Product

Resources

About