Efficient Sliding Window Computation for NN-Based Template Matching

Nearest neighbor search is a fundamental problem in computer vision and machine learning. The straightforward solution, linear scan, is both computationally and memory intensive in large scale high-dimensional cases, hence is not preferable in practice. Therefore, there have been a lot of interests in algorithms that perform approximate nearest neighbor (ANN) search. In this paper, we propose a novel addition-based vector quantization algorithm, Asymmetric Mapping Quantization (AMQ), to efficiently conduct ANN search. Unlike existing addition-based quantization methods that suffer from handling the problem caused by the norm of database vector, we map the query vector and database vector using different mapping functions to transform the computation of L-2 distance to inner product similarity, thus do not need to evaluate the norm of database vector. Moreover, we further propose Distributed Asymmetric Mapping Quantization (DAMQ) to enable AMQ to work on very large dataset by distributed learning. Extensive experiments on approximate nearest neighbor search and image retrieval validate the merits of the proposed AMQ and DAMQ.

The Perspective-3-Point problem (P3P) is a classical and fundamental problem in computer vision. All possible solution sets for the P3P problem are from 1 to 4 solutions. In this paper, we propose a very simple way to reduce the ambiguity of numbers of possible solutions in P3P using a planar mirror. For three reference points, if they and their reflections in a planar mirror are both observed, we may obtain two P3P problems: One is from the three original reference points, and the other is from their reflections. A trivial procedure may be suggested: Solve for each of the two P3P problems, and then find the intersections of the two solution sets. Different from the trivial case, we propose an efficient method which employs the ratio relations of the unknowns in the two P3P problems. The ratio relations are arise from mirror reflection, and can be easily determined before solving the two P3P problems. With the ratio relations, a system of 6 equations with 3 unknowns can be determined. To solve the over-constraint problem, we utilize an efficient algorithm by finding all local minima of least-squares residual. Experiments validate our approach.

In the areas of computer vision and computer graphics, analyzing information about three-dimensional objects from two-dimensional images is essential. Shape from shading is a very fundamental problem in computer vision, which uses features and patterns in images to infer the shape of the object in our viewing direction. The influence of solving this problem is significant. For example, a typical application of shape from shading is in astronomy, in which we can obtain landscape features of a planet by analyzing the information from the image we get.

The nearest neighbor search is an essential operation for many computer vision, data mining, and machine learning problems. Since it is so widely used, the nearest neighbor search should be as fast as possible. This paper explores lower bound-based approaches to speed up the exact nearest neighbor search in high dimensional Euclidean space. We compute the lower bound of Euclidean Distance by using the block vectors and Cauchy–Schwartz inequality. The proposed lower bound is calculated efficiently and is close to the real Euclidean Distance. Besides, the preprocessing step of the proposal has linear time complexity. Given a query, during the procedure of identifying the nearest neighbor, our method can eliminate many expensive actual distance computations using the lower bound to approximate Euclidean Distance. In addition, we develop a multilevel lower bound strategy, which calculates the lower bound step by step and utilizes the multistep filtering mechanism to improve the searching process further. Theoretical analysis is provided to show that the proposals can guarantee to obtain the same result as the brute-force search. Comprehensive experiments on 16 public data sets collected from various domains demonstrate that our approach performs well in finding the exact nearest neighbor compared to related competitors. The experimental results also illustrate that the multilevel lower bound strategy is effective.

In this paper we provide new fast and simple solutions to two important minimal problems in computer vision, the five-point relative pose problem and the six-point focal length problem. We show that these two problems can easily be formulated as polynomial eigenvalue problems of degree three and two and solved using standard efficient numerical algorithms. Our solutions are somewhat more stable than state-of-the-art solutions by Nister and Stewenius and are in some sense more straightforward and easier to implement since polynomial eigenvalue problems are well studied with many efficient and robust algorithms available. The quality of the solvers is demonstrated in experiments 1.

The task of matching image patches is a fundamental problem in computer vision. When sufficiently textured patches are normalized up to similarity transformation, a simple Normalized Cross Correlation (NCC) of corresponding patches will give a high value. In practice, using it on patches per se may not perform well due to the noisy variations of pixel intensities. A more prudent approach will be to apply it to the abstract features extracted by a deep convolutional network. We study the applicability of an NCC based convolutional network for the task of Patch Matching. Further, there may be cases where the network may fail due to insufficient textures. In those cases, a simple pixel difference based method will be beneficial. To this end, we propose to improve the two basic architectures, Siamese networks and Central-Surround stream networks, using robust matching layers for learning the similarities of patches, assisted by a simple cross-entropy loss function. We empirically verify the performance of the proposed models on the challenging UBC Patches dataset and show that they are close to the state-of-the-art. Further, we evaluate their resilience to large illumination changes in two experimental scenarios: 1) by manually varying the patches of UBC Patches by an affine model 2) by using the publicly available Webcam dataset. We demonstrate that our models are indeed very resilient to illumination variations; they reduce the false positive rate to nearly 10%, and improve over the popular methods by nearly 5%. Further, we demonstrate the generalisability of the proposed NCC based matching layer by applying it to Face Recognition and show that it improves the performances of well known networks on a real-world, surveillance dataset.

Previous chapter Next chapter Full AccessProceedings Proceedings of the 2020 ACM-SIAM Symposium on Discrete Algorithms (SODA)Faster Deterministic and Las Vegas Algorithms for Offline Approximate Nearest Neighbors in High DimensionsJosh Alman, Timothy M. Chan, and Ryan WilliamsJosh Alman, Timothy M. Chan, and Ryan Williamspp.637 - 649Chapter DOI:https://doi.org/10.1137/1.9781611975994.39PDFBibTexSections ToolsAdd to favoritesExport CitationTrack CitationsEmail SectionsAboutAbstract We present a deterministic, truly subquadratic algorithm for offline (1 + ε)-approximate nearest or farthest neighbor search (in particular, the closest pair or diameter problem) in Hamming space in any dimension d ≤ nδ, for a sufficiently small constant δ > 0. The running time of the algorithm is roughly for nearest neighbors, or for farthest. The algorithm follows from a simple combination of expander walks, Chebyshev polynomials, and rectangular matrix multiplication. We also show how to eliminate errors in the previous Monte Carlo randomized algorithm of Alman, Chan, and Williams [FOCS’16] for offline approximate nearest or farthest neighbors, and obtain a Las Vegas randomized algorithm with expected running time . Finally, we note a simplification of Alman, Chan, and Williams' method and obtain a slightly improved Monte Carlo randomized algorithm with running time . As one application, we obtain improved deterministic and randomized (1 + ε)-approximation algorithms for MAX-SAT. Previous chapter Next chapter RelatedDetails Published:2020eISBN:978-1-61197-599-4 https://doi.org/10.1137/1.9781611975994Book Series Name:ProceedingsBook Code:PRDA20Book Pages:xxii + 3011

Image classification is a fundamental problem in computer vision, which deserves much attention in the past decade. To provide a comprehensive recall for the image classification task based on the deep learning algorithms, this paper first provides a brief history of computer vision and convolutional neural networks. Then, the current state of research and the direction of development of deep learning-based convolutional neural network models for image classification are examined. In addition, the basic model structure, convolution feature extraction, and pooling operations of standard and convolutional neural networks are also introduced. This paper summarizes the development of convolutional neural network models in recent years. In summary, based on the review of the development of image classification algorithms at home and abroad, the current mainstream image classification algorithms and frontier progress are summarized and analyzed, and the existing problems and future development directions of image classification are summarized and prospected. It can be concluded that deep learning-based methods has a great potential in this case.

Kernel cross-view collaborative representation based classification for person re-identification

Visual object tracking is a classical, but still open research problem in computer vision, with many real world applications. The problem is challenging due to several factors, such as illumination variation, occlusions, camera motion and appearance changes. Such problems can be alleviated by constructing robust, discriminative and computationally efficient visual features. Recently, biologically-inspired channel representations [9] have shown to provide promising results in many applications ranging from autonomous driving to visual tracking.

Visual tracking is a challenging problem in computer vision. Most state-of-the-art visual trackers are lack of an ability to adapt affine transforms of objects. The Lucas-Kanade(L-K) algorithm is a classic gradient decent search method which has been widely applied in template matching, image alignment and visual tracking. Despite the L-K algorithm has a good property on adapting targets affine transforms, it always falls into a local optimal since the background information is accumulatively introduced to the template. In this paper, we propose a novel approach, called the Re-located L-K algorithm(RL-K), to handle affine transforms of objects. In particular, we utilize the Multiple Channel Correlation Filters(MMCFs) to deal with the translations of the object before the L-K's gradient decent search step. In addition, we integrate the gradient orientation features into the L-K algorithm to improve its robustness to visual tracking. The experimental results show that, comparing with the state-of-the-art affine tracking approaches, our approach is running at real-time without accuracy losing.

Labelling the lines of a planar line drawing of a 3-D object in a way that reflects the geometric properties of the object is a much studied problem in computer vision and considered to be an important step towards understanding the object from its 2-D drawing. Combinatorially, the labellability problem is a restricted version of the Constraint Satisfaction Problem and has been shown to be NP-complete even for images of polyhedral scenes. In this paper, we examine scenes that consist of a set of objects each obtained by rotating a polygon around an arbitrary axis. The objects are allowed to arbitrarily intersect or overlay. We show that for these scenes, there is a sequential linear-time labelling algorithm. Moreover, we show that the algorithm has a fast parallel version that executes in O(log4 n) time on a EREW PRAM with O(n 3/log2 n) processors. The algorithm not only answers the decision problem of labellability, but also produces a legal labelling, if there is one. This parallel algorithm should be contrasted with the techniques of dealing with special cases of the constraint satisfaction problem. These techniques employ an effective, but inherently sequential, relaxation procedure in order to restrict the domains of the variables.

We propose a novel computationally efficient correspondence algorithm to identify the correspondence of similar features of the Reference frame to Search frame in the dynamic image sequence analysis for face tracking. The correspondence of the features between the set of images is the central problem in Computer Vision(CV), Image Analysis(IA) and Pattern Recognition(PR). In these areas Correspondence of features is the combinatorial explosion problem or NP-hard, because of their exhaustive searching for features in the sequence of image frames. This correspondence has three major steps such as segmentation, feature extraction and matching. In this paper we propose block-based segmentation in which the frame which contains the face to be track can be segmented as window of size 20*20 which covers the face. Here we consider the window trace mean is the feature since this is rotation invariant and it covers all the rows and columns of the window and it reduces the dimensionality and finally for matching we used the minimum absolute difference method which does not involves more computations since no multiplication operation is involved. We restrict the searching space to [-3,+3] pixels horizontally and vertically in the search image. Experimental results show that this novel algorithm could achieve much higher computational reduction as compared with full search (FS), Diamond search (DS)[6], Cross Diamond Search (CDS)[7,8] and Area Based Correspondence[14] algorithm for face tracking image sequence while similar prediction accuracy is maintained and it is especially suitable for video conferencing and slow moving dynamic image sequence.

Labelling the lines of a planar line drawing of a 3-D object in a way that reflects the geometric properties of the object is a much studied problem in computer vision, considered to be an important step towards understanding the object from its 2-D drawing. Combinatorially, the labellability problem is a Constraint Satisfaction Problem and has been shown to be NP-complete even for images of polyhedral scenes. In this paper, we examine scenes that consist of a set of objects each obtained by rotating a polygon around an arbitrary axis. The objects are allowed to arbitrarily intersect or overlay. We show that for these scenes, there is a sequential lineartime labelling algorithm. Moreover, we show that the algorithm has a fast parallel version that executes inO(log3 n) time on an Exclusive-Read-Exclusive-Write Parallel Random Access Machine withO(n 3/log3 n) processors. The algorithm not only answers the decision problem of labellability, but also produces a legal labelling, if there is one. This parallel algorithm should be contrasted with the techniques of dealing with special cases of the constraint satisfaction problem. These techniques employ an effective, but inherently sequential, relaxation procedure in order to restrict the domains of the variables.

The domain of existentially uncertain spatial data refers to objects that are modelled using an existential probability accompanying spatial data values. An interesting and challenging query type over existentially uncertain data is the search of the nearest neighbor (NN), since the probability of a potential dataset object to be the NN of the query object depends on the locations and probabilities of other points in the same dataset. In this paper, following a statistical approach, we estimate the average number of the NNsrequired to answer probabilistic thresholding NN(PTNN) queries as function of the threshold t, allowing us to utilize existing approaches and propose a cost model for such queries. Based on the same statistical approach, we propose an efficient algorithm for PTNN queries over arbitrarily structured existentially uncertain spatial data. Our experimental study demonstrates the accuracy and efficiency of the proposed techniques.