Geometric min-Hashing: Finding a (thick) needle in a haystack

Fast retrieval methods are critical for large-scale and data-driven vision applications. Recent work has explored ways to embed high-dimensional features or complex distance functions into a low-dimensional Hamming space where items can be efficiently searched. However, existing methods do not apply for high-dimensional kernelized data when the underlying feature embedding for the kernel is unknown. We show how to generalize locality-sensitive hashing to accommodate arbitrary kernel functions, making it possible to preserve the algorithm's sub-linear time similarity search guarantees for a wide class of useful similarity functions. Since a number of successful image-based kernels have unknown or incomputable embeddings, this is especially valuable for image retrieval tasks. We validate our technique on several large-scale datasets, and show that it enables accurate and fast performance for example-based object classification, feature matching, and content-based retrieval.

Proceedings of the seventeen ACM international conference on Multimedia - MM '09, 2009

An efficient indexing method is essential for content-based image retrieval with the exponential growth in large-scale videos and photos. Recently, hash-based methods (e.g., locality sensitive hashing -LSH) have been shown efficient for similarity search. We extend such hash-based methods for retrieving images represented by bags of (high-dimensional) feature points. Though promising, the hash-based image object search suffers from low recall rates. To boost the hash-based search quality, we propose two novel expansion strategies -intra-expansion and inter-expansion. The former expands more target feature points similar to those in the query and the latter mines those feature points that shall co-occur with the search targets but not present in the query. We further exploit variations for the proposed methods. Experimenting in two consumer-photo benchmarks, we will show that the proposed expansion methods are complementary to each other and can collaboratively contribute up to 76.3% (average) relative improvement over the original hash-based method.

It is known that relative feature location is important in representing objects, but assumptions that make learning tractable often simplify how structure is encoded e.g. spatial pooling or star models. For example, techniques such as spatial pyramid matching (SPM), in-conjunction with machine learning techniques perform well. However, there are limitations to such spatial encoding schemes which discard important information about the layout of features. In contrast, we propose to use the object itself to choose the basis of the features in an object centric approach. In doing so we return to the early work of geometric hashing but demonstrate how such approaches can be scaled-up to modern day object detection challenges in terms of both the number of examples and their variability. We apply a two stage process; initially filtering background features to localise the objects and then hashing the remaining pairwise features in an affine invariant model. During learning, we identify class-wise key feature predictors. We validate our detection and classification of objects on the PASCAL VOC'07 and '11 and CarDb datasets and compare with state of the art detectors and classifiers. Importantly we demonstrate how structure in features can be efficiently identified and how its inclusion can increase performance. This feature centric learning technique allows us to localise objects even without object annotation during training and the resultant segmentation provides accurate state of the art object localization, without the need for annotations.

Abstract: Geometric Hashing is a well-known technique for object recognition. This paper proposes a novel method aimed at improving the performance of Geometric Hashing in terms of robustness toward occlusion and clutter. To this purpose, it employs feature descriptors to notably decrease the amount of false positives that generally arise under these conditions.

Computer Vision—ECCV'96, 1996

We address the problem of 3D object recognition from a single 2D image using a model database. We develop a new method called enhanced geometric hashing. This approach allows us to solve for the indexing and the matching problem in one pass with linear complexity. Use of quasi-invariants allows us to index images in a new type of geometric hashing table. They include topological information of the observed objects inducing a high numerical stability.

2011

Abstract—We present a scalable image retrieval system based jointly on text annotations and visual content. Previous approaches in content based image retrieval often suffer from the semantic gap problem and long retrieving time. The solution that we propose aims at resolving these two issues by indexing and retrieving images using both their text descriptions and visual content, such as features in colour, texture and shape. A query in this system consists of keywords, a sample image and relevant parameters.

Proc. BMVC2007, 2007

The geometric hashing (GH) is a well-known model-based object recognition technique with good properties both in retrieval speed and required amount of memory. However, it has a significant weak point; as the number of objects increases, both retrieval speed and required amount ...

IEEE Computational Science and Engineering, 1997

In this paper we describe the Geometric Hashing paradigm for matching of a set of geometric features against a database of such feature sets. Speci c examples are model based object recognition in computer vision for which this technique was originally developed, matching of volumetric data obtained from CT or MRI images of di erent persons, matching of an individual ngerprint versus a database, matching the molecular surface of a receptor molecule against a data base of drugs and so on. The features considered can be points, segments, in nite lines, corners and any other geometric entities. The matching is performed under any non-elastic geometric transformation such as the rigid, similarity, a ne and projective transformations in any dimension. Moreover, the problem addressed is the much more di cult partial matching problem, where one tries to detect (previously unknown) large subsets of the feature set which are compatible with subsets of the database feature sets. The technique is highly e cient and is of low polynomial complexity in the feature set. The power of the polynomial depends on the class of transformations that the feature sets are allowed to undergo. The e ciency is achieved by an indexing scheme which preserves the geometric rigidity constraints of a shape, thus distinguishing this approach from standard indexing techniques which exploit local transformation invariants. The technique is rst described in the simple case of a 2-D similarity transformation. Then its extensions are discussed and a weighted voting scheme is presented. This scheme allows to reformulate the technique in the Bayesian framework.

Proceedings of the AAAI Conference on Artificial Intelligence

Hashing is a popular approximate nearest neighbor search approach for large-scale image retrieval. Supervised hashing, which incorporates similarity/dissimilarity information on entity pairs to improve the quality of hashing function learning, has recently received increasing attention. However, in the existing supervised hashing methods for images, an input image is usually encoded by a vector of hand-crafted visual features. Such hand-crafted feature vectors do not necessarily preserve the accurate semantic similarities of images pairs, which may often degrade the performance of hashing function learning. In this paper, we propose a supervised hashing method for image retrieval, in which we automatically learn a good image representation tailored to hashing as well as a set of hash functions. The proposed method has two stages. In the first stage, given the pairwise similarity matrix $S$ over training images, we propose a scalable coordinate descent method to decompose $S$ into a ...

IEEE Computational Science and Engineering, 1997