@inproceedings{2948, abstract = {Many visual datasets are traditionally used to analyze the performance of different learning techniques. The evaluation is usually done within each dataset, therefore it is questionable if such results are a reliable indicator of true generalization ability. We propose here an algorithm to exploit the existing data resources when learning on a new multiclass problem. Our main idea is to identify an image representation that decomposes orthogonally into two subspaces: a part specific to each dataset, and a part generic to, and therefore shared between, all the considered source sets. This allows us to use the generic representation as un-biased reference knowledge for a novel classification task. By casting the method in the multi-view setting, we also make it possible to use different features for different databases. We call the algorithm MUST, Multitask Unaligned Shared knowledge Transfer. Through extensive experiments on five public datasets, we show that MUST consistently improves the cross-datasets generalization performance.}, author = {Tommasi, Tatiana and Quadrianto, Novi and Caputo, Barbara and Lampert, Christoph}, location = {Daejeon, Korea}, pages = {1 -- 15}, publisher = {Springer}, title = {{Beyond dataset bias: Multi-task unaligned shared knowledge transfer}}, doi = {10.1007/978-3-642-37331-2_1}, volume = {7724}, year = {2013}, } @misc{3321, author = {Quadrianto, Novi and Lampert, Christoph}, booktitle = {Encyclopedia of Systems Biology}, editor = {Dubitzky, Werner and Wolkenhauer, Olaf and Cho, Kwang and Yokota, Hiroki}, pages = {1069 -- 1069}, publisher = {Springer}, title = {{Kernel based learning}}, doi = {10.1007/978-1-4419-9863-7_604}, volume = {3}, year = {2013}, } @inproceedings{2293, abstract = {Many computer vision problems have an asymmetric distribution of information between training and test time. In this work, we study the case where we are given additional information about the training data, which however will not be available at test time. This situation is called learning using privileged information (LUPI). We introduce two maximum-margin techniques that are able to make use of this additional source of information, and we show that the framework is applicable to several scenarios that have been studied in computer vision before. Experiments with attributes, bounding boxes, image tags and rationales as additional information in object classification show promising results.}, author = {Sharmanska, Viktoriia and Quadrianto, Novi and Lampert, Christoph}, location = {Sydney, Australia}, pages = {825 -- 832}, publisher = {IEEE}, title = {{Learning to rank using privileged information}}, doi = {10.1109/ICCV.2013.107}, year = {2013}, } @inproceedings{2294, abstract = {In this work we propose a system for automatic classification of Drosophila embryos into developmental stages. While the system is designed to solve an actual problem in biological research, we believe that the principle underly- ing it is interesting not only for biologists, but also for researchers in computer vision. The main idea is to combine two orthogonal sources of information: one is a classifier trained on strongly invariant features, which makes it applicable to images of very different conditions, but also leads to rather noisy predictions. The other is a label propagation step based on a more powerful similarity measure that however is only consistent within specific subsets of the data at a time. In our biological setup, the information sources are the shape and the staining patterns of embryo images. We show experimentally that while neither of the methods can be used by itself to achieve satisfactory results, their combina- tion achieves prediction quality comparable to human performance.}, author = {Kazmar, Tomas and Kvon, Evgeny and Stark, Alexander and Lampert, Christoph}, location = {Sydney, Australia}, publisher = {IEEE}, title = {{Drosophila Embryo Stage Annotation using Label Propagation}}, doi = {10.1109/ICCV.2013.139}, year = {2013}, } @article{2516, abstract = {We study the problem of object recognition for categories for which we have no training examples, a task also called zero-data or zero-shot learning. This situation has hardly been studied in computer vision research, even though it occurs frequently: the world contains tens of thousands of different object classes and for only few of them image collections have been formed and suitably annotated. To tackle the problem we introduce attribute-based classification: objects are identified based on a high-level description that is phrased in terms of semantic attributes, such as the object's color or shape. Because the identification of each such property transcends the specific learning task at hand, the attribute classifiers can be pre-learned independently, e.g. from existing image datasets unrelated to the current task. Afterwards, new classes can be detected based on their attribute representation, without the need for a new training phase. In this paper we also introduce a new dataset, Animals with Attributes, of over 30,000 images of 50 animal classes, annotated with 85 semantic attributes. Extensive experiments on this and two more datasets show that attribute-based classification indeed is able to categorize images without access to any training images of the target classes.}, author = {Lampert, Christoph and Nickisch, Hannes and Harmeling, Stefan}, journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence}, number = {3}, pages = {453 -- 465}, publisher = {IEEE}, title = {{Attribute-based classification for zero-shot learning of object categories}}, doi = {10.1109/TPAMI.2013.140}, volume = {36}, year = {2013}, } @inproceedings{2520, abstract = {We propose a probabilistic model to infer supervised latent variables in the Hamming space from observed data. Our model allows simultaneous inference of the number of binary latent variables, and their values. The latent variables preserve neighbourhood structure of the data in a sense that objects in the same semantic concept have similar latent values, and objects in different concepts have dissimilar latent values. We formulate the supervised infinite latent variable problem based on an intuitive principle of pulling objects together if they are of the same type, and pushing them apart if they are not. We then combine this principle with a flexible Indian Buffet Process prior on the latent variables. We show that the inferred supervised latent variables can be directly used to perform a nearest neighbour search for the purpose of retrieval. We introduce a new application of dynamically extending hash codes, and show how to effectively couple the structure of the hash codes with continuously growing structure of the neighbourhood preserving infinite latent feature space.}, author = {Quadrianto, Novi and Sharmanska, Viktoriia and Knowles, David and Ghahramani, Zoubin}, booktitle = {Proceedings of the 29th conference uncertainty in Artificial Intelligence}, isbn = {9780974903996}, location = {Bellevue, WA, United States}, pages = {527 -- 536}, publisher = {AUAI Press}, title = {{The supervised IBP: Neighbourhood preserving infinite latent feature models}}, year = {2013}, } @inproceedings{2825, abstract = {We study the problem of maximum marginal prediction (MMP) in probabilistic graphical models, a task that occurs, for example, as the Bayes optimal decision rule under a Hamming loss. MMP is typically performed as a two-stage procedure: one estimates each variable's marginal probability and then forms a prediction from the states of maximal probability. In this work we propose a simple yet effective technique for accelerating MMP when inference is sampling-based: instead of the above two-stage procedure we directly estimate the posterior probability of each decision variable. This allows us to identify the point of time when we are sufficiently certain about any individual decision. Whenever this is the case, we dynamically prune the variables we are confident about from the underlying factor graph. Consequently, at any time only samples of variables whose decision is still uncertain need to be created. Experiments in two prototypical scenarios, multi-label classification and image inpainting, show that adaptive sampling can drastically accelerate MMP without sacrificing prediction accuracy.}, author = {Lampert, Christoph}, location = {Lake Tahoe, NV, United States}, pages = {82 -- 90}, publisher = {Neural Information Processing Systems Foundation}, title = {{Dynamic pruning of factor graphs for maximum marginal prediction}}, volume = {1}, year = {2012}, } @inproceedings{2915, author = {Kroemer, Oliver and Lampert, Christoph and Peters, Jan}, publisher = {Deutsches Zentrum für Luft und Raumfahrt}, title = {{Multi-modal learning for dynamic tactile sensing}}, year = {2012}, } @inproceedings{3124, abstract = {We consider the problem of inference in a graphical model with binary variables. While in theory it is arguably preferable to compute marginal probabilities, in practice researchers often use MAP inference due to the availability of efficient discrete optimization algorithms. We bridge the gap between the two approaches by introducing the Discrete Marginals technique in which approximate marginals are obtained by minimizing an objective function with unary and pairwise terms over a discretized domain. This allows the use of techniques originally developed for MAP-MRF inference and learning. We explore two ways to set up the objective function - by discretizing the Bethe free energy and by learning it from training data. Experimental results show that for certain types of graphs a learned function can outperform the Bethe approximation. We also establish a link between the Bethe free energy and submodular functions. }, author = {Korc, Filip and Kolmogorov, Vladimir and Lampert, Christoph}, location = {Edinburgh, Scotland}, publisher = {ICML}, title = {{Approximating marginals using discrete energy minimization}}, year = {2012}, } @inproceedings{3125, abstract = {We propose a new learning method to infer a mid-level feature representation that combines the advantage of semantic attribute representations with the higher expressive power of non-semantic features. The idea lies in augmenting an existing attribute-based representation with additional dimensions for which an autoencoder model is coupled with a large-margin principle. This construction allows a smooth transition between the zero-shot regime with no training example, the unsupervised regime with training examples but without class labels, and the supervised regime with training examples and with class labels. The resulting optimization problem can be solved efficiently, because several of the necessity steps have closed-form solutions. Through extensive experiments we show that the augmented representation achieves better results in terms of object categorization accuracy than the semantic representation alone.}, author = {Sharmanska, Viktoriia and Quadrianto, Novi and Lampert, Christoph}, location = {Florence, Italy}, number = {PART 5}, pages = {242 -- 255}, publisher = {Springer}, title = {{Augmented attribute representations}}, doi = {10.1007/978-3-642-33715-4_18}, volume = {7576}, year = {2012}, } @inproceedings{3126, abstract = {In this work we propose a new information-theoretic clustering algorithm that infers cluster memberships by direct optimization of a non-parametric mutual information estimate between data distribution and cluster assignment. Although the optimization objective has a solid theoretical foundation it is hard to optimize. We propose an approximate optimization formulation that leads to an efficient algorithm with low runtime complexity. The algorithm has a single free parameter, the number of clusters to find. We demonstrate superior performance on several synthetic and real datasets. }, author = {Müller, Andreas and Nowozin, Sebastian and Lampert, Christoph}, location = {Graz, Austria}, pages = {205 -- 215}, publisher = {Springer}, title = {{Information theoretic clustering using minimal spanning trees}}, doi = {10.1007/978-3-642-32717-9_21}, volume = {7476}, year = {2012}, } @inproceedings{3127, abstract = {When searching for characteristic subpatterns in potentially noisy graph data, it appears self-evident that having multiple observations would be better than having just one. However, it turns out that the inconsistencies introduced when different graph instances have different edge sets pose a serious challenge. In this work we address this challenge for the problem of finding maximum weighted cliques. We introduce the concept of most persistent soft-clique. This is subset of vertices, that 1) is almost fully or at least densely connected, 2) occurs in all or almost all graph instances, and 3) has the maximum weight. We present a measure of clique-ness, that essentially counts the number of edge missing to make a subset of vertices into a clique. With this measure, we show that the problem of finding the most persistent soft-clique problem can be cast either as: a) a max-min two person game optimization problem, or b) a min-min soft margin optimization problem. Both formulations lead to the same solution when using a partial Lagrangian method to solve the optimization problems. By experiments on synthetic data and on real social network data, we show that the proposed method is able to reliably find soft cliques in graph data, even if that is distorted by random noise or unreliable observations.}, author = {Quadrianto, Novi and Lampert, Christoph and Chen, Chao}, booktitle = {Proceedings of the 29th International Conference on Machine Learning}, location = {Edinburgh, United Kingdom}, pages = {211--218}, publisher = {ML Research Press}, title = {{The most persistent soft-clique in a set of sampled graphs}}, year = {2012}, } @article{3164, abstract = {Overview of the Special Issue on structured prediction and inference.}, author = {Blaschko, Matthew and Lampert, Christoph}, journal = {International Journal of Computer Vision}, number = {3}, pages = {257 -- 258}, publisher = {Springer}, title = {{Guest editorial: Special issue on structured prediction and inference}}, doi = {10.1007/s11263-012-0530-y}, volume = {99}, year = {2012}, } @article{3248, abstract = {We describe RTblob, a high speed vision system that detects objects in cluttered scenes based on their color and shape at a speed of over 800 frames/s. Because the system is available as open-source software and relies only on off-the-shelf PC hardware components, it can provide the basis for multiple application scenarios. As an illustrative example, we show how RTblob can be used in a robotic table tennis scenario to estimate ball trajectories through 3D space simultaneously from four cameras images at a speed of 200 Hz.}, author = {Lampert, Christoph and Peters, Jan}, issn = {1861-8219}, journal = {Journal of Real-Time Image Processing}, number = {1}, pages = {31 -- 41}, publisher = {Springer}, title = {{Real-time detection of colored objects in multiple camera streams with off-the-shelf hardware components}}, doi = {10.1007/s11554-010-0168-3}, volume = {7}, year = {2012}, } @misc{5396, abstract = {We consider the problem of inference in agraphical model with binary variables. While in theory it is arguably preferable to compute marginal probabilities, in practice researchers often use MAP inference due to the availability of efficient discrete optimization algorithms. We bridge the gap between the two approaches by introducing the Discrete Marginals technique in which approximate marginals are obtained by minimizing an objective function with unary and pair-wise terms over a discretized domain. This allows the use of techniques originally devel-oped for MAP-MRF inference and learning. We explore two ways to set up the objective function - by discretizing the Bethe free energy and by learning it from training data. Experimental results show that for certain types of graphs a learned function can out-perform the Bethe approximation. We also establish a link between the Bethe free energy and submodular functions.}, author = {Korc, Filip and Kolmogorov, Vladimir and Lampert, Christoph}, issn = {2664-1690}, pages = {13}, publisher = {IST Austria}, title = {{Approximating marginals using discrete energy minimization}}, doi = {10.15479/AT:IST-2012-0003}, year = {2012}, } @inproceedings{3163, abstract = {We study multi-label prediction for structured output sets, a problem that occurs, for example, in object detection in images, secondary structure prediction in computational biology, and graph matching with symmetries. Conventional multilabel classification techniques are typically not applicable in this situation, because they require explicit enumeration of the label set, which is infeasible in case of structured outputs. Relying on techniques originally designed for single-label structured prediction, in particular structured support vector machines, results in reduced prediction accuracy, or leads to infeasible optimization problems. In this work we derive a maximum-margin training formulation for multi-label structured prediction that remains computationally tractable while achieving high prediction accuracy. It also shares most beneficial properties with single-label maximum-margin approaches, in particular formulation as a convex optimization problem, efficient working set training, and PAC-Bayesian generalization bounds.}, author = {Lampert, Christoph}, location = {Granada, Spain}, publisher = {Neural Information Processing Systems Foundation}, title = {{Maximum margin multi-label structured prediction}}, year = {2011}, } @inproceedings{3319, abstract = {We address the problem of metric learning for multi-view data, namely the construction of embedding projections from data in different representations into a shared feature space, such that the Euclidean distance in this space provides a meaningful within-view as well as between-view similarity. Our motivation stems from the problem of cross-media retrieval tasks, where the availability of a joint Euclidean distance function is a pre-requisite to allow fast, in particular hashing-based, nearest neighbor queries. We formulate an objective function that expresses the intuitive concept that matching samples are mapped closely together in the output space, whereas non-matching samples are pushed apart, no matter in which view they are available. The resulting optimization problem is not convex, but it can be decomposed explicitly into a convex and a concave part, thereby allowing efficient optimization using the convex-concave procedure. Experiments on an image retrieval task show that nearest-neighbor based cross-view retrieval is indeed possible, and the proposed technique improves the retrieval accuracy over baseline techniques.}, author = {Quadrianto, Novi and Lampert, Christoph}, location = {Bellevue, United States}, pages = {425 -- 432}, publisher = {ML Research Press}, title = {{Learning multi-view neighborhood preserving projections}}, year = {2011}, } @article{3320, abstract = {Powerful statistical models that can be learned efficiently from large amounts of data are currently revolutionizing computer vision. These models possess a rich internal structure reflecting task-specific relations and constraints. This monograph introduces the reader to the most popular classes of structured models in computer vision. Our focus is discrete undirected graphical models which we cover in detail together with a description of algorithms for both probabilistic inference and maximum a posteriori inference. We discuss separately recently successful techniques for prediction in general structured models. In the second part of this monograph we describe methods for parameter learning where we distinguish the classic maximum likelihood based methods from the more recent prediction-based parameter learning methods. We highlight developments to enhance current models and discuss kernelized models and latent variable models. To make the monograph more practical and to provide links to further study we provide examples of successful application of many methods in the computer vision literature.}, author = {Nowozin, Sebastian and Lampert, Christoph}, journal = {Foundations and Trends in Computer Graphics and Vision}, number = {3-4}, pages = {185 -- 365}, publisher = {Now Publishers}, title = {{Structured learning and prediction in computer vision}}, doi = {10.1561/0600000033}, volume = {6}, year = {2011}, } @misc{3322, abstract = {We study multi-label prediction for structured output spaces, a problem that occurs, for example, in object detection in images, secondary structure prediction in computational biology, and graph matching with symmetries. Conventional multi-label classification techniques are typically not applicable in this situation, because they require explicit enumeration of the label space, which is infeasible in case of structured outputs. Relying on techniques originally designed for single- label structured prediction, in particular structured support vector machines, results in reduced prediction accuracy, or leads to infeasible optimization problems. In this work we derive a maximum-margin training formulation for multi-label structured prediction that remains computationally tractable while achieving high prediction accuracy. It also shares most beneficial properties with single-label maximum-margin approaches, in particular a formulation as a convex optimization problem, efficient working set training, and PAC-Bayesian generalization bounds.}, author = {Lampert, Christoph}, booktitle = {NIPS: Neural Information Processing Systems}, publisher = {Neural Information Processing Systems Foundation}, title = {{Maximum margin multi label structured prediction}}, year = {2011}, } @inproceedings{3336, abstract = {We introduce TopoCut: a new way to integrate knowledge about topological properties (TPs) into random field image segmentation model. Instead of including TPs as additional constraints during minimization of the energy function, we devise an efficient algorithm for modifying the unary potentials such that the resulting segmentation is guaranteed with the desired properties. Our method is more flexible in the sense that it handles more topology constraints than previous methods, which were only able to enforce pairwise or global connectivity. In particular, our method is very fast, making it for the first time possible to enforce global topological properties in practical image segmentation tasks.}, author = {Chen, Chao and Freedman, Daniel and Lampert, Christoph}, booktitle = {CVPR: Computer Vision and Pattern Recognition}, isbn = {978-1-4577-0394-2}, location = {Colorado Springs, CO, United States}, pages = {2089 -- 2096}, publisher = {IEEE}, title = {{Enforcing topological constraints in random field image segmentation}}, doi = {10.1109/CVPR.2011.5995503}, year = {2011}, }