Explore the vast range of titles available.

The α -expansion algorithm has had a significant impact in computer vision due to its generality, effectiveness, and speed. It is commonly used to minimize energies that involve unary, pairwise, and specialized higher-order terms. Our main algorithmic contribution is an extension of α -expansion that also optimizes “label costs” with well-characterized optimality bounds. Label costs penalize a solution based on the set of labels that appear in it, for example by simply penalizing the number of labels in the solution. Our energy has a natural interpretation as minimizing description length (MDL) and sheds light on classical algorithms like K -means and expectation-maximization (EM). Label costs are useful for multi-model fitting and we demonstrate several such applications: homography detection, motion segmentation, image segmentation, and compression. Our C++ and MATLAB code is publicly available http://vision.csd.uwo.ca/code/ .

Geometric model fitting is a typical chicken-&-egg problem: data points should be clustered based on geometric proximity to models whose unknown parameters must be estimated at the same time. Most existing methods, including generalizations of RANSAC , greedily search for models with most inliers (within a threshold) ignoring overall classification of points. We formulate geometric multi-model fitting as an optimal labeling problem with a global energy function balancing geometric errors and regularity of inlier clusters. Regularization based on spatial coherence (on some near-neighbor graph) and/or label costs is NP hard. Standard combinatorial algorithms with guaranteed approximation bounds (e.g. α -expansion) can minimize such regularization energies over a finite set of labels, but they are not directly applicable to a continuum of labels, e.g. in line fitting. Our proposed approach ( PEaRL ) combines model sampling from data points as in RANSAC with iterative re-estimation of inliers and models’ parameters based on a global regularization functional. This technique efficiently explores the continuum of labels in the context of energy minimization. In practice, PEaRL converges to a good quality local minimum of the energy automatically selecting a small number of models that best explain the whole data set. Our tests demonstrate that our energy-based approach significantly improves the current state of the art in geometric model fitting currently dominated by various greedy generalizations of RANSAC .

Feature matching and model fitting are fundamental problems in multi-view geometry. They are chicken-&-egg problems: if models are known it is easier to find matches and vice versa. Standard multi-view geometry techniques sequentially solve feature matching and model fitting as two independent problems after making fairly restrictive assumptions. For example, matching methods rely on strong discriminative power of feature descriptors, which fail for stereo images with repetitive textures or wide baseline. Also, model fitting methods assume given feature matches, which are not known a priori. Moreover, when data supports multiple models the fitting problem becomes challenging even with known matches and current methods commonly use heuristics. One of the main contributions of this thesis is a joint formulation of fitting and matching problems. We are first to introduce an objective function combining both matching and multi-model estimation. We also propose an approximation algorithm for the corresponding NP-hard optimization problem using block-coordinate descent with respect to matching and model fitting variables. For fixed models, our method uses min-cost-max-flow based algorithm to solve a generalization of a linear assignment problem with label cost (sparsity constraint). Fixed matching case reduces to multi-model fitting subproblem, which is interesting in its own right. In contrast to standard heuristic approaches, we introduce global objective functions for multi-model fitting using various forms of regularization (spatial smoothness and sparsity) and propose a graph-cut based optimization algorithm, PEaRL. Experimental results show that our proposed mathematical formulations and optimization algorithms improve the accuracy and robustness of model estimation over the state-of-the-art in computer vision.

While 3D Gaussian Splatting has recently become popular for neural rendering, current methods rely on carefully engineered cloning and splitting strategies for placing Gaussians, which can lead to poor-quality renderings, and reliance on a good initialization. In this work, we rethink the set of 3D Gaussians as a random sample drawn from an underlying probability distribution describing the physical representation of the scene-in other words, Markov Chain Monte Carlo (MCMC) samples. Under this view, we show that the 3D Gaussian updates can be converted as Stochastic Gradient Langevin Dynamics (SGLD) updates by simply introducing noise. We then rewrite the densification and pruning strategies in 3D Gaussian Splatting as simply a deterministic state transition of MCMC samples, removing these heuristics from the framework. To do so, we revise the 'cloning' of Gaussians into a relocalization scheme that approximately preserves sample probability. To encourage efficient use of Gaussians, we introduce a regularizer that promotes the removal of unused Gaussians. On various standard evaluation scenes, we show that our method provides improved rendering quality, easy control over the number of Gaussians, and robustness to initialization.

Text-to-image diffusion models are now capable of generating images that are often indistinguishable from real images. To generate such images, these models must understand the semantics of the objects they are asked to generate. In this work we show that, without any training, one can leverage this semantic knowledge within diffusion models to find semantic correspondences - locations in multiple images that have the same semantic meaning. Specifically, given an image, we optimize the prompt embeddings of these models for maximum attention on the regions of interest. These optimized embeddings capture semantic information about the location, which can then be transferred to another image. By doing so we obtain results on par with the strongly supervised state of the art on the PF-Willow dataset and significantly outperform (20.9% relative for the SPair-71k dataset) any existing weakly or unsupervised method on PF-Willow, CUB-200 and SPair-71k datasets.

We present an approach to accelerate Neural Field training by efficiently selecting sampling locations. While Neural Fields have recently become popular, it is often trained by uniformly sampling the training domain, or through handcrafted heuristics. We show that improved convergence and final training quality can be achieved by a soft mining technique based on importance sampling: rather than either considering or ignoring a pixel completely, we weigh the corresponding loss by a scalar. To implement our idea we use Langevin Monte-Carlo sampling. We show that by doing so, regions with higher error are being selected more frequently, leading to more than 2x improvement in convergence speed. The code and related resources for this study are publicly available at https://ubc-vision.github.io/nf-soft-mining/.

While 3D Gaussian Splatting has recently become popular for neural rendering, current methods rely on carefully engineered cloning and splitting strategies for placing Gaussians, which can lead to poor-quality renderings, and reliance on a good initialization. In this work, we rethink the set of 3D Gaussians as a random sample drawn from an underlying probability distribution describing the physical representation of the scene-in other words, Markov Chain Monte Carlo (MCMC) samples. Under this view, we show that the 3D Gaussian updates can be converted as Stochastic Gradient Langevin Dynamics (SGLD) updates by simply introducing noise. We then rewrite the densification and pruning strategies in 3D Gaussian Splatting as simply a deterministic state transition of MCMC samples, removing these heuristics from the framework. To do so, we revise the 'cloning' of Gaussians into a relocalization scheme that approximately preserves sample probability. To encourage efficient use of Gaussians, we introduce a regularizer that promotes the removal of unused Gaussians. On various standard evaluation scenes, we show that our method provides improved rendering quality, easy control over the number of Gaussians, and robustness to initialization.

Unsupervised learning of keypoints and landmarks has seen significant progress with the help of modern neural network architectures, but performance is yet to match the supervised counterpart, making their practicability questionable. We leverage the emergent knowledge within text-to-image diffusion models, towards more robust unsupervised keypoints. Our core idea is to find text embeddings that would cause the generative model to consistently attend to compact regions in images (i.e. keypoints). To do so, we simply optimize the text embedding such that the cross-attention maps within the denoising network are localized as Gaussians with small standard deviations. We validate our performance on multiple datasets: the CelebA, CUB-200-2011, Tai-Chi-HD, DeepFashion, and Human3.6m datasets. We achieve significantly improved accuracy, sometimes even outperforming supervised ones, particularly for data that is non-aligned and less curated. Our code is publicly available and can be found through our project page: https://ubc-vision.github.io/StableKeypoints/

Many standard approaches for geometric model fitting are based on pre-matched image features. Typically, such pre-matching uses only feature appearances (e.g. SIFT) and a large number of non-unique features must be discarded in order to control the false positive rate. In contrast, we solve feature matching and multi-model fitting problems in a joint optimization framework. This paper proposes several fit-&-match energy formulations based on a generalization of the assignment problem. We developed an efficient solver based on min-cost-max-flow algorithm that finds near optimal solutions. Our approach significantly increases the number of detected matches. In practice, energy-based joint fitting & matching allows to increase the distance between view-points previously restricted by robustness of local SIFT-matching and to improve the model fitting accuracy when compared to state-of-the-art multi-model fitting techniques.

We propose a novel efficient and lightweight model for human pose estimation from a single image. Our model is designed to achieve competitive results at a fraction of the number of parameters and computational cost of various state-of-the-art methods. To this end, we explicitly incorporate part-based structural and geometric priors in a hierarchical prediction framework. At the coarsest resolution, and in a manner similar to classical part-based approaches, we leverage the kinematic structure of the human body to propagate convolutional feature updates between the keypoints or body parts. Unlike classical approaches, we adopt end-to-end training to learn this geometric prior through feature updates from data. We then propagate the feature representation at the coarsest resolution up the hierarchy to refine the predicted pose in a coarse-to-fine fashion. The final network effectively models the geometric prior and intuition within a lightweight deep neural network, yielding state-of-the-art results for a model of this size on two standard datasets, Leeds Sports Pose and MPII Human Pose.