Paper Group ANR 372
Optimisation of photometric stereo methods by non-convex variational minimisation. How hard is it to cross the room? – Training (Recurrent) Neural Networks to steer a UAV. Entropy, neutro-entropy and anti-entropy for neutrosophic information. Cross-lingual Distillation for Text Classification. Inception Score, Label Smoothing, Gradient Vanishing a …
Optimisation of photometric stereo methods by non-convex variational minimisation
| Title | Optimisation of photometric stereo methods by non-convex variational minimisation |
| Authors | Georg Radow, Laurent Hoeltgen, Yvain Quéau, Michael Breuß |
| Abstract | Estimating shape and appearance of a three dimensional object from a given set of images is a classic research topic that is still actively pursued. Among the various techniques available, PS is distinguished by the assumption that the underlying input images are taken from the same point of view but under different lighting conditions. The most common techniques provide the shape information in terms of surface normals. In this work, we instead propose to minimise a much more natural objective function, namely the reprojection error in terms of depth. Minimising the resulting non-trivial variational model for PS allows to recover the depth of the photographed scene directly. As a solving strategy, we follow an approach based on a recently published optimisation scheme for non-convex and non-smooth cost functions. The main contributions of our paper are of theoretical nature. A technical novelty in our framework is the usage of matrix differential calculus. We supplement our approach by a detailed convergence analysis of the resulting optimisation algorithm and discuss possibilities to ease the computational complexity. At hand of an experimental evaluation we discuss important properties of the method. Overall, our strategy achieves more accurate results than competing approaches. The experiments also highlights some practical aspects of the underlying optimisation algorithm that may be of interest in a more general context. |
| Tasks | |
| Published | 2017-09-29 |
| URL | http://arxiv.org/abs/1709.10437v1 |
| http://arxiv.org/pdf/1709.10437v1.pdf | |
| PWC | https://paperswithcode.com/paper/optimisation-of-photometric-stereo-methods-by |
| Repo | |
| Framework | |
How hard is it to cross the room? – Training (Recurrent) Neural Networks to steer a UAV
| Title | How hard is it to cross the room? – Training (Recurrent) Neural Networks to steer a UAV |
| Authors | Klaas Kelchtermans, Tinne Tuytelaars |
| Abstract | This work explores the feasibility of steering a drone with a (recurrent) neural network, based on input from a forward looking camera, in the context of a high-level navigation task. We set up a generic framework for training a network to perform navigation tasks based on imitation learning. It can be applied to both aerial and land vehicles. As a proof of concept we apply it to a UAV (Unmanned Aerial Vehicle) in a simulated environment, learning to cross a room containing a number of obstacles. So far only feedforward neural networks (FNNs) have been used to train UAV control. To cope with more complex tasks, we propose the use of recurrent neural networks (RNN) instead and successfully train an LSTM (Long-Short Term Memory) network for controlling UAVs. Vision based control is a sequential prediction problem, known for its highly correlated input data. The correlation makes training a network hard, especially an RNN. To overcome this issue, we investigate an alternative sampling method during training, namely window-wise truncated backpropagation through time (WW-TBPTT). Further, end-to-end training requires a lot of data which often is not available. Therefore, we compare the performance of retraining only the Fully Connected (FC) and LSTM control layers with networks which are trained end-to-end. Performing the relatively simple task of crossing a room already reveals important guidelines and good practices for training neural control networks. Different visualizations help to explain the behavior learned. |
| Tasks | Imitation Learning |
| Published | 2017-02-24 |
| URL | http://arxiv.org/abs/1702.07600v1 |
| http://arxiv.org/pdf/1702.07600v1.pdf | |
| PWC | https://paperswithcode.com/paper/how-hard-is-it-to-cross-the-room-training |
| Repo | |
| Framework | |
Entropy, neutro-entropy and anti-entropy for neutrosophic information
| Title | Entropy, neutro-entropy and anti-entropy for neutrosophic information |
| Authors | Vasile Patrascu |
| Abstract | This approach presents a multi-valued representation of the neutrosophic information. It highlights the link between the bifuzzy information and neutrosophic one. The constructed deca-valued structure shows the neutrosophic information complexity. This deca-valued structure led to construction of two new concepts for the neutrosophic information: neutro-entropy and anti-entropy. These two concepts are added to the two existing: entropy and non-entropy. Thus, we obtained the following triad: entropy, neutro-entropy and anti-entropy. |
| Tasks | |
| Published | 2017-06-18 |
| URL | http://arxiv.org/abs/1706.05643v1 |
| http://arxiv.org/pdf/1706.05643v1.pdf | |
| PWC | https://paperswithcode.com/paper/entropy-neutro-entropy-and-anti-entropy-for |
| Repo | |
| Framework | |
Cross-lingual Distillation for Text Classification
| Title | Cross-lingual Distillation for Text Classification |
| Authors | Ruochen Xu, Yiming Yang |
| Abstract | Cross-lingual text classification(CLTC) is the task of classifying documents written in different languages into the same taxonomy of categories. This paper presents a novel approach to CLTC that builds on model distillation, which adapts and extends a framework originally proposed for model compression. Using soft probabilistic predictions for the documents in a label-rich language as the (induced) supervisory labels in a parallel corpus of documents, we train classifiers successfully for new languages in which labeled training data are not available. An adversarial feature adaptation technique is also applied during the model training to reduce distribution mismatch. We conducted experiments on two benchmark CLTC datasets, treating English as the source language and German, French, Japan and Chinese as the unlabeled target languages. The proposed approach had the advantageous or comparable performance of the other state-of-art methods. |
| Tasks | Model Compression, Text Classification |
| Published | 2017-05-05 |
| URL | http://arxiv.org/abs/1705.02073v2 |
| http://arxiv.org/pdf/1705.02073v2.pdf | |
| PWC | https://paperswithcode.com/paper/cross-lingual-distillation-for-text |
| Repo | |
| Framework | |
Inception Score, Label Smoothing, Gradient Vanishing and -log(D(x)) Alternative
| Title | Inception Score, Label Smoothing, Gradient Vanishing and -log(D(x)) Alternative |
| Authors | Zhiming Zhou, Weinan Zhang, Jun Wang |
| Abstract | In this article, we mathematically study several GAN related topics, including Inception score, label smoothing, gradient vanishing and the -log(D(x)) alternative. — An advanced version is included in arXiv:1703.02000 “Activation Maximization Generative Adversarial Nets”. Please refer Section 6 in 1703.02000 for detailed analysis on Inception Score, and refer its appendix for the discussions on Label Smoothing, Gradient Vanishing and -log(D(x)) Alternative. |
| Tasks | |
| Published | 2017-08-05 |
| URL | http://arxiv.org/abs/1708.01729v3 |
| http://arxiv.org/pdf/1708.01729v3.pdf | |
| PWC | https://paperswithcode.com/paper/inception-score-label-smoothing-gradient |
| Repo | |
| Framework | |
U-Phylogeny: Undirected Provenance Graph Construction in the Wild
| Title | U-Phylogeny: Undirected Provenance Graph Construction in the Wild |
| Authors | Aparna Bharati, Daniel Moreira, Allan Pinto, Joel Brogan, Kevin Bowyer, Patrick Flynn, Walter Scheirer, Anderson Rocha |
| Abstract | Deriving relationships between images and tracing back their history of modifications are at the core of Multimedia Phylogeny solutions, which aim to combat misinformation through doctored visual media. Nonetheless, most recent image phylogeny solutions cannot properly address cases of forged composite images with multiple donors, an area known as multiple parenting phylogeny (MPP). This paper presents a preliminary undirected graph construction solution for MPP, without any strict assumptions. The algorithm is underpinned by robust image representative keypoints and different geometric consistency checks among matching regions in both images to provide regions of interest for direct comparison. The paper introduces a novel technique to geometrically filter the most promising matches as well as to aid in the shared region localization task. The strength of the approach is corroborated by experiments with real-world cases, with and without image distractors (unrelated cases). |
| Tasks | graph construction |
| Published | 2017-05-31 |
| URL | http://arxiv.org/abs/1705.11187v1 |
| http://arxiv.org/pdf/1705.11187v1.pdf | |
| PWC | https://paperswithcode.com/paper/u-phylogeny-undirected-provenance-graph |
| Repo | |
| Framework | |
The Geodesic Distance between $\mathcal{G}_I^0$ Models and its Application to Region Discrimination
| Title | The Geodesic Distance between $\mathcal{G}_I^0$ Models and its Application to Region Discrimination |
| Authors | José Naranjo-Torres, Juliana Gambini, Alejandro C. Frery |
| Abstract | The $\mathcal{G}_I^0$ distribution is able to characterize different regions in monopolarized SAR imagery. It is indexed by three parameters: the number of looks (which can be estimated in the whole image), a scale parameter and a texture parameter. This paper presents a new proposal for feature extraction and region discrimination in SAR imagery, using the geodesic distance as a measure of dissimilarity between $\mathcal{G}_I^0$ models. We derive geodesic distances between models that describe several practical situations, assuming the number of looks known, for same and different texture and for same and different scale. We then apply this new tool to the problems of (i)~identifying edges between regions with different texture, and (ii)~quantify the dissimilarity between pairs of samples in actual SAR data. We analyze the advantages of using the geodesic distance when compared to stochastic distances. |
| Tasks | |
| Published | 2017-01-01 |
| URL | http://arxiv.org/abs/1701.00294v1 |
| http://arxiv.org/pdf/1701.00294v1.pdf | |
| PWC | https://paperswithcode.com/paper/the-geodesic-distance-between-mathcalg_i0 |
| Repo | |
| Framework | |
Learning Deep Neural Network Representations for Koopman Operators of Nonlinear Dynamical Systems
| Title | Learning Deep Neural Network Representations for Koopman Operators of Nonlinear Dynamical Systems |
| Authors | Enoch Yeung, Soumya Kundu, Nathan Hodas |
| Abstract | The Koopman operator has recently garnered much attention for its value in dynamical systems analysis and data-driven model discovery. However, its application has been hindered by the computational complexity of extended dynamic mode decomposition; this requires a combinatorially large basis set to adequately describe many nonlinear systems of interest, e.g. cyber-physical infrastructure systems, biological networks, social systems, and fluid dynamics. Often the dictionaries generated for these problems are manually curated, requiring domain-specific knowledge and painstaking tuning. In this paper we introduce a deep learning framework for learning Koopman operators of nonlinear dynamical systems. We show that this novel method automatically selects efficient deep dictionaries, outperforming state-of-the-art methods. We benchmark this method on partially observed nonlinear systems, including the glycolytic oscillator and show it is able to predict quantitatively 100 steps into the future, using only a single timepoint, and qualitative oscillatory behavior 400 steps into the future. |
| Tasks | |
| Published | 2017-08-22 |
| URL | http://arxiv.org/abs/1708.06850v2 |
| http://arxiv.org/pdf/1708.06850v2.pdf | |
| PWC | https://paperswithcode.com/paper/learning-deep-neural-network-representations |
| Repo | |
| Framework | |
CASP Solutions for Planning in Hybrid Domains
| Title | CASP Solutions for Planning in Hybrid Domains |
| Authors | Marcello Balduccini, Daniele Magazzeni, Marco Maratea, Emily LeBlanc |
| Abstract | CASP is an extension of ASP that allows for numerical constraints to be added in the rules. PDDL+ is an extension of the PDDL standard language of automated planning for modeling mixed discrete-continuous dynamics. In this paper, we present CASP solutions for dealing with PDDL+ problems, i.e., encoding from PDDL+ to CASP, and extensions to the algorithm of the EZCSP CASP solver in order to solve CASP programs arising from PDDL+ domains. An experimental analysis, performed on well-known linear and non-linear variants of PDDL+ domains, involving various configurations of the EZCSP solver, other CASP solvers, and PDDL+ planners, shows the viability of our solution. |
| Tasks | |
| Published | 2017-04-12 |
| URL | http://arxiv.org/abs/1704.03574v2 |
| http://arxiv.org/pdf/1704.03574v2.pdf | |
| PWC | https://paperswithcode.com/paper/casp-solutions-for-planning-in-hybrid-domains |
| Repo | |
| Framework | |
Neural System Combination for Machine Translation
| Title | Neural System Combination for Machine Translation |
| Authors | Long Zhou, Wenpeng Hu, Jiajun Zhang, Chengqing Zong |
| Abstract | Neural machine translation (NMT) becomes a new approach to machine translation and generates much more fluent results compared to statistical machine translation (SMT). However, SMT is usually better than NMT in translation adequacy. It is therefore a promising direction to combine the advantages of both NMT and SMT. In this paper, we propose a neural system combination framework leveraging multi-source NMT, which takes as input the outputs of NMT and SMT systems and produces the final translation. Extensive experiments on the Chinese-to-English translation task show that our model archives significant improvement by 5.3 BLEU points over the best single system output and 3.4 BLEU points over the state-of-the-art traditional system combination methods. |
| Tasks | Machine Translation |
| Published | 2017-04-21 |
| URL | http://arxiv.org/abs/1704.06393v1 |
| http://arxiv.org/pdf/1704.06393v1.pdf | |
| PWC | https://paperswithcode.com/paper/neural-system-combination-for-machine |
| Repo | |
| Framework | |
Image Stitching by Line-guided Local Warping with Global Similarity Constraint
| Title | Image Stitching by Line-guided Local Warping with Global Similarity Constraint |
| Authors | Tian-Zhu Xiang, Gui-Song Xia, Xiang Bai, Liangpei Zhang |
| Abstract | Low-textured image stitching remains a challenging problem. It is difficult to achieve good alignment and it is easy to break image structures due to insufficient and unreliable point correspondences. Moreover, because of the viewpoint variations between multiple images, the stitched images suffer from projective distortions. To solve these problems, this paper presents a line-guided local warping method with a global similarity constraint for image stitching. Line features which serve well for geometric descriptions and scene constraints, are employed to guide image stitching accurately. On one hand, the line features are integrated into a local warping model through a designed weight function. On the other hand, line features are adopted to impose strong geometric constraints, including line correspondence and line colinearity, to improve the stitching performance through mesh optimization. To mitigate projective distortions, we adopt a global similarity constraint, which is integrated with the projective warps via a designed weight strategy. This constraint causes the final warp to slowly change from a projective to a similarity transformation across the image. Finally, the images undergo a two-stage alignment scheme that provides accurate alignment and reduces projective distortion. We evaluate our method on a series of images and compare it with several other methods. The experimental results demonstrate that the proposed method provides a convincing stitching performance and that it outperforms other state-of-the-art methods. |
| Tasks | Image Stitching |
| Published | 2017-02-25 |
| URL | http://arxiv.org/abs/1702.07935v2 |
| http://arxiv.org/pdf/1702.07935v2.pdf | |
| PWC | https://paperswithcode.com/paper/image-stitching-by-line-guided-local-warping |
| Repo | |
| Framework | |
MagNet: a Two-Pronged Defense against Adversarial Examples
| Title | MagNet: a Two-Pronged Defense against Adversarial Examples |
| Authors | Dongyu Meng, Hao Chen |
| Abstract | Deep learning has shown promising results on hard perceptual problems in recent years. However, deep learning systems are found to be vulnerable to small adversarial perturbations that are nearly imperceptible to human. Such specially crafted perturbations cause deep learning systems to output incorrect decisions, with potentially disastrous consequences. These vulnerabilities hinder the deployment of deep learning systems where safety or security is important. Attempts to secure deep learning systems either target specific attacks or have been shown to be ineffective. In this paper, we propose MagNet, a framework for defending neural network classifiers against adversarial examples. MagNet does not modify the protected classifier or know the process for generating adversarial examples. MagNet includes one or more separate detector networks and a reformer network. Different from previous work, MagNet learns to differentiate between normal and adversarial examples by approximating the manifold of normal examples. Since it does not rely on any process for generating adversarial examples, it has substantial generalization power. Moreover, MagNet reconstructs adversarial examples by moving them towards the manifold, which is effective for helping classify adversarial examples with small perturbation correctly. We discuss the intrinsic difficulty in defending against whitebox attack and propose a mechanism to defend against graybox attack. Inspired by the use of randomness in cryptography, we propose to use diversity to strengthen MagNet. We show empirically that MagNet is effective against most advanced state-of-the-art attacks in blackbox and graybox scenarios while keeping false positive rate on normal examples very low. |
| Tasks | |
| Published | 2017-05-25 |
| URL | http://arxiv.org/abs/1705.09064v2 |
| http://arxiv.org/pdf/1705.09064v2.pdf | |
| PWC | https://paperswithcode.com/paper/magnet-a-two-pronged-defense-against |
| Repo | |
| Framework | |
Quasi-homography warps in image stitching
| Title | Quasi-homography warps in image stitching |
| Authors | Nan Li, Yifang Xu, Chao Wang |
| Abstract | The naturalness of warps is gaining extensive attentions in image stitching. Recent warps such as SPHP and AANAP, use global similarity warps to mitigate projective distortion (which enlarges regions), however, they necessarily bring in perspective distortion (which generates inconsistencies). In this paper, we propose a novel quasi-homography warp, which effectively balances the perspective distortion against the projective distortion in the non-overlapping region to create a more natural-looking panorama. Our approach formulates the warp as the solution of a bivariate system, where perspective distortion and projective distortion are characterized as slope preservation and scale linearization respectively. Because our proposed warp only relies on a global homography, thus it is totally parameter-free. A comprehensive experiment shows that a quasi-homography warp outperforms some state-of-the-art warps in urban scenes, including homography, AutoStitch and SPHP. A user study demonstrates that it wins most users’ favor, comparing to homography and SPHP. |
| Tasks | Image Stitching |
| Published | 2017-01-27 |
| URL | http://arxiv.org/abs/1701.08006v2 |
| http://arxiv.org/pdf/1701.08006v2.pdf | |
| PWC | https://paperswithcode.com/paper/quasi-homography-warps-in-image-stitching |
| Repo | |
| Framework | |
Synonym Discovery with Etymology-based Word Embeddings
| Title | Synonym Discovery with Etymology-based Word Embeddings |
| Authors | Seunghyun Yoon, Pablo Estrada, Kyomin Jung |
| Abstract | We propose a novel approach to learn word embeddings based on an extended version of the distributional hypothesis. Our model derives word embedding vectors using the etymological composition of words, rather than the context in which they appear. It has the strength of not requiring a large text corpus, but instead it requires reliable access to etymological roots of words, making it specially fit for languages with logographic writing systems. The model consists on three steps: (1) building an etymological graph, which is a bipartite network of words and etymological roots, (2) obtaining the biadjacency matrix of the etymological graph and reducing its dimensionality, (3) using columns/rows of the resulting matrices as embedding vectors. We test our model in the Chinese and Sino-Korean vocabularies. Our graphs are formed by a set of 117,000 Chinese words, and a set of 135,000 Sino-Korean words. In both cases we show that our model performs well in the task of synonym discovery. |
| Tasks | Word Embeddings |
| Published | 2017-09-29 |
| URL | http://arxiv.org/abs/1709.10445v2 |
| http://arxiv.org/pdf/1709.10445v2.pdf | |
| PWC | https://paperswithcode.com/paper/synonym-discovery-with-etymology-based-word |
| Repo | |
| Framework | |
RIPML: A Restricted Isometry Property based Approach to Multilabel Learning
| Title | RIPML: A Restricted Isometry Property based Approach to Multilabel Learning |
| Authors | Akshay Soni, Yashar Mehdad |
| Abstract | The multilabel learning problem with large number of labels, features, and data-points has generated a tremendous interest recently. A recurring theme of these problems is that only a few labels are active in any given datapoint as compared to the total number of labels. However, only a small number of existing work take direct advantage of this inherent extreme sparsity in the label space. By the virtue of Restricted Isometry Property (RIP), satisfied by many random ensembles, we propose a novel procedure for multilabel learning known as RIPML. During the training phase, in RIPML, labels are projected onto a random low-dimensional subspace followed by solving a least-square problem in this subspace. Inference is done by a k-nearest neighbor (kNN) based approach. We demonstrate the effectiveness of RIPML by conducting extensive simulations and comparing results with the state-of-the-art linear dimensionality reduction based approaches. |
| Tasks | Dimensionality Reduction |
| Published | 2017-02-16 |
| URL | http://arxiv.org/abs/1702.05181v1 |
| http://arxiv.org/pdf/1702.05181v1.pdf | |
| PWC | https://paperswithcode.com/paper/ripml-a-restricted-isometry-property-based |
| Repo | |
| Framework | |
