In this paper, we explore methods for learning local image descriptors from training data. We describe a set of building blocks for constructing descriptors which can be combined together and jointly optimized so as to minimize the error of a nearest-neighbor classifier. We consider both linear and nonlinear transforms with dimensionality reduction, and make use of discriminant learning techniques such as Linear Discriminant Analysis (LDA) and Powell minimization to solve for the parameters. Using these techniques, we obtain descriptors that exceed state-of-the-art performance with low dimensionality. In addition to new experiments and recommendations for descriptor learning, we are also making available a new and realistic ground truth data set based on multiview stereo data.

Local image descriptors that are highly discriminative, computational efficient, and with low storage footprint have long been a dream goal of computer vision research. In this paper, we focus on learning such descriptors, which make use of the DAISY configuration and are simple to compute both sparsely and densely. We develop a new training set of match/non-match image patches which improves on previous work. We test a wide variety of gradient and steerable filter based configurations and optimize over all parameters to obtain low matching errors for the descriptors. We further explore robust normalization, dimension reduction and dynamic range reduction to increase the discriminative power and yet reduce the storage requirement of the learned descriptors. All these enable us to obtain highly efficient local descriptors: e.g, 13.2% error at 13 bytes storage per descriptor, compared with 26.1% error at 128 bytes for SIFT.

Web image search is difficult in part because a handful of keywords are generally insufficient for characterizing the visual properties of an image. Popular engines have begun to provide tags based on simple characteristics of images (such as tags for black and white images or images that contain a face), but such approaches are limited by the fact that it is unclear what tags end users want to be able to use in examining Web image search results. This paper presents CueFlik, a Web image search application that allows end users to quickly create their own rules for re ranking images based on their visual characteristics. End users can then re rank any future Web image search results according to their rule. In an experiment we present in this paper, end users quickly create effective rules for such concepts as "product photos", "portraits of people", and "clipart". When asked to conceive of and create their own rules, participants create such rules as "sports action shot" with images from queries for "basketball" and "football". CueFlik represents both a promising new approach to Web image search and an important study in end user interactive machine learning.

In this paper we study interest point descriptors for image matching and 3D reconstruction. We examine the building blocks of descriptor algorithms and evaluate numerous combinations of components. Various published descriptors such as SIFT, GLOH, and Spin images can be cast into our framework. For each candidate algorithm we learn good choices for parameters using a training set consisting of patches from a multi-image 3D reconstruction where accurate ground-truth matches are known. The best descriptors were those with log polar histogramming regions and feature vectors constructed from rectified outputs of steerable quadrature filters. At a 95% detection rate these gave one third of the incorrect matches produced by SIFT.

Invariant feature descriptors such as SIFT and GLOH have been demonstrated to be very robust for image matching and visual recognition. However, such descriptors are generally parameterised in very high dimensional spaces e.g. 128 dimensions in the case of SIFT. This limits the performance of feature matching techniques in terms of speed and scalability. Furthermore, these descriptors have traditionally been carefully hand crafted by manually tuning many parameters. In this paper, we tackle both of these problems by formulating descriptor design as a non- parametric dimensionality reduction problem. In contrast to previous approaches that use only the global statistics of the inputs, we adopt a discriminative approach. Starting from a large training set of labelled match/non-match pairs, we pursue lower dimensional embeddings that are optimised for their discriminative power. Extensive comparative experiments demonstrate that we can exceed the performance of the current state of the art techniques such as SIFT with far fewer dimensions, and with virtually no parameters to be tuned by hand.

This paper describes a novel multi-view matching framework based on a new type of invariant feature. Our features are located at Harris corners in discrete scale-space and oriented using a blurred local gradient. This defines a rotationally invariant frame in which we sample a feature descriptor, which consists of an 8 /spl times/ 8 patch of bias/gain normalised intensity values. The density of features in the image is controlled using a novel adaptive non-maximal suppression algorithm, which gives a better spatial distribution of features than previous approaches. Matching is achieved using a fast nearest neighbour algorithm that indexes features based on their low frequency Haar wavelet coefficients. We also introduce a novel outlier rejection procedure that verifies a pairwise feature match based on a background distribution of incorrect feature matches. Feature matches are refined using RANSAC and used in an automatic 2D panorama stitcher that has been extensively tested on hundreds of sample inputs.

The ability to interactively control viewpoint while watching a video is an exciting application of image-based rendering. The goal of our work is to render dynamic scenes with interactive viewpoint control using a relatively small number of video cameras. In this paper, we show how high-quality video-based rendering of dynamic scenes can be accomplished using multiple synchronized video streams combined with novel image-based modeling and rendering algorithms. Once these video streams have been processed, we can synthesize any intermediate view between cameras at any time, with the potential for space-time manipulation.In our approach, we first use a novel color segmentation-based stereo algorithm to generate high-quality photoconsistent correspondences across all camera views. Mattes for areas near depth discontinuities are then automatically extracted to reduce artifacts during view synthesis. Finally, a novel temporal two-layer compressed representation that handles matting is developed for rendering at interactive rates.

Typical video footage captured using an off-the-shelf camcorder suffers from limited dynamic range. This paper describes our approach to generate high dynamic range (HDR) video from an image sequence of a dynamic scene captured while rapidly varying the exposure of each frame. Our approach consists of three parts: automatic exposure control during capture, HDR stitching across neighboring frames, and tonemapping for viewing. HDR stitching requires accurately registering neighboring frames and choosing appropriate pixels for computing the radiance map. We show examples for a variety of dynamic scenes. We also show how we can compensate for scene and camera movement when creating an HDR still from a series of bracketed still photographs.