Refine
Year of publication
- 2018 (13) (remove)
Document Type
- Conference Proceeding (9)
- Article (3)
- Doctoral Thesis (1)
Language
- English (13)
Keywords
- Classification (1)
- Deep learning (1)
- Imaging (1)
- Learning to cluster (1)
- Perceptual grouping (1)
- Recurrent neural network (1)
- Screening (1)
- Speaker clustering (1)
- Speaker recognition (1)
- Speech & image clustering (1)
Institute
- Institut für Optische Systeme - IOS (13) (remove)
Know when you don't know
(2018)
Deep convolutional neural networks show outstanding performance in image-based phenotype classification given that all existing phenotypes are presented during the training of the network. However, in real-world high-content screening (HCS) experiments, it is often impossible to know all phenotypes in advance. Moreover, novel phenotype discovery itself can be an HCS outcome of interest. This aspect of HCS is not yet covered by classical deep learning approaches. When presenting an image with a novel phenotype to a trained network, it fails to indicate a novelty discovery but assigns the image to a wrong phenotype. To tackle this problem and address the need for novelty detection, we use a recently developed Bayesian approach for deep neural networks called Monte Carlo (MC) dropout to define different uncertainty measures for each phenotype prediction. With real HCS data, we show that these uncertainty measures allow us to identify novel or unclear phenotypes. In addition, we also found that the MC dropout method results in a significant improvement of classification accuracy. The proposed procedure used in our HCS case study can be easily transferred to any existing network architecture and will be beneficial in terms of accuracy and novelty detection.
Simon Grimm examines new multi-microphone signal processing strategies that aim to achieve noise reduction and dereverberation. Therefore, narrow-band signal enhancement approaches are combined with broad-band processing in terms of directivity based beamforming. Previously introduced formulations of the multichannel Wiener filter rely on the second order statistics of the speech and noise signals. The author analyses how additional knowledge about the location of a speaker as well as the microphone arrangement can be used to achieve further noise reduction and dereverberation.
Optical surface inspection: A novelty detection approach based on CNN-encoded texture features
(2018)
In inspection systems for textured surfaces, a reference texture is typically known before novel examples are inspected. Mostly, the reference is only available in a digital format. As a consequence, there is no dataset of defective examples available that could be used to train a classifier. We propose a texture model approach to novelty detection. The texture model uses features encoded by a convolutional neural network (CNN) trained on natural image data. The CNN activations represent the specific characteristics of the digital reference texture which are learned by a one-class classifier. We evaluate our novelty detector in a digital print inspection scenario. The inspection unit is based on a camera array and a flashing light illumination which allows for inline capturing of multichannel images at a high rate. In order to compare our results to manual inspection, we integrated our inspection unit into an industrial single-pass printing system.
Deep neural networks have been successfully applied to problems such as image segmentation, image super-resolution, coloration and image inpainting. In this work we propose the use of convolutional neural networks (CNN) for image inpainting of large regions in high-resolution textures. Due to limited computational resources processing high-resolution images with neural networks is still an open problem. Existing methods separate inpainting of global structure and the transfer of details, which leads to blurry results and loss of global coherence in the detail transfer step. Based on advances in texture synthesis using CNNs we propose patch-based image inpainting by a single network topology that is able to optimize for global as well as detail texture statistics. Our method is capable of filling large inpainting regions, oftentimes exceeding quality of comparable methods for images of high-resolution (2048x2048px). For reference patch look-up we propose to use the same summary statistics that are used in the inpainting process.
In this paper we present a method using deep learning to compute parametrizations for B-spline curve approximation. Existing methods consider the computation of parametric values and a knot vector as separate problems. We propose to train interdependent deep neural networks to predict parametric values and knots. We show that it is possible to include B-spline curve approximation directly into the neural network architecture. The resulting parametrizations yield tight approximations and are able to outperform state-of-the-art methods.
Knot placement for curve approximation is a well known and yet open problem in geometric modeling. Selecting knot values that yield good approximations is a challenging task, based largely on heuristics and user experience. More advanced approaches range from parametric averaging to genetic algorithms.
In this paper, we propose to use Support Vector Machines (SVMs) to determine suitable knot vectors for B-spline curve approximation. The SVMs are trained to identify locations in a sequential point cloud where knot placement will improve the approximation error. After the training phase, the SVM can assign, to each point set location, a so-called score. This score is based on geometric and differential geometric features of points. It measures the quality of each location to be used as knots in the subsequent approximation. From these scores, the final knot vector can be constructed exploring the topography of the score-vector without the need for iteration or optimization in the approximation process. Knot vectors computed with our approach outperform state of the art methods and yield tighter approximations.
We propose a novel end-to-end neural network architecture that, once trained, directly outputs a probabilistic clustering of a batch of input examples in one pass. It estimates a distribution over the number of clusters k, and for each 1≤k≤kmax, a distribution over the individual cluster assignment for each data point. The network is trained in advance in a supervised fashion on separate data to learn grouping by any perceptual similarity criterion based on pairwise labels (same/different group). It can then be applied to different data containing different groups. We demonstrate promising performance on high-dimensional data like images (COIL-100) and speech (TIMIT). We call this “learning to cluster” and show its conceptual difference to deep metric learning, semi-supervise clustering and other related approaches while having the advantage of performing learnable clustering fully end-to-end.
Algorithms for calculating the string edit distance are used in e.g. information retrieval and document analysis systems or for evaluation of text recognizers. Text recognition based on CTC-trained LSTM networks includes a decoding step to produce a string, possibly using a language model, and evaluation using the string edit distance. The decoded string can further be used as a query for database search, e.g. in document retrieval. We propose to closely integrate dictionary search with text recognition to train both combined in a continuous fashion. This work shows that LSTM networks are capable of calculating the string edit distance while allowing for an exchangeable dictionary to separate learned algorithm from data. This could be a step towards integrating text recognition and dictionary search in one deep network.
Visualization-Assisted Development of Deep Learning Models in Offline Handwriting Recognition
(2018)
Deep learning is a field of machine learning that has been the focus of active research and successful applications in recent years. Offline handwriting recognition is one of the research fields and applications were deep neural networks have shown high accuracy. Deep learning models and their training pipeline show a large amount of hyper-parameters in their data selection, transformation, network topology and training process that are sometimes interdependent. This increases the overall difficulty and time necessary for building and training a model for a specific data set and task at hand. This work proposes a novel visualization-assisted workflow that guides the model developer through the hyper-parameter search in order to identify relevant parameters and modify them in a meaningful way. This decreases the overall time necessary for building and training a model. The contributions of this work are a workflow for hyper-parameter search in offline handwriting recognition and a heat map based visualization technique for deep neural networks in multi-line offline handwriting recognition. This work applies to offline handwriting recognition, but the general workflow can possibly be adapted to other tasks as well.
Offline handwriting recognition systems often use LSTM networks, trained with line- or word-images. Multi-line text makes it necessary to use segmentation to explicitly obtain these images. Skewed, curved, overlapping, incorrectly written text, or noise can lead to errors during segmentation of multi-line text and reduces the overall recognition capacity of the system. Last year has seen the introduction of deep learning methods capable of segmentation-free recognition of whole paragraphs. Our method uses Conditional Random Fields to represent text and align it with the network output to calculate a loss function for training. Experiments are promising and show that the technique is capable of training a LSTM multi-line text recognition system.
Deep neural networks have become a veritable alternative to classic speaker recognition and clustering methods in recent years. However, while the speech signal clearly is a time series, and despite the body of literature on the benefits of prosodic (suprasegmental) features, identifying voices has usually not been approached with sequence learning methods. Only recently has a recurrent neural network (RNN) been successfully applied to this task, while the use of convolutional neural networks (CNNs) (that are not able to capture arbitrary time dependencies, unlike RNNs) still prevails. In this paper, we show the effectiveness of RNNs for speaker recognition by improving state of the art speaker clustering performance and robustness on the classic TIMIT benchmark. We provide arguments why RNNs are superior by experimentally showing a “sweet spot” of the segment length for successfully capturing prosodic information that has been theoretically predicted in previous work.