Pau Riba, Anjan Dutta, Lutz Goldmann, Alicia Fornes, Oriol Ramos Terrades, & Josep Llados. (2019). Table Detection in Invoice Documents by Graph Neural Networks. In 15th International Conference on Document Analysis and Recognition (pp. 122–127).
Abstract: Tabular structures in documents offer a complementary dimension to the raw textual data, representing logical or quantitative relationships among pieces of information. In digital mail room applications, where a large amount of
administrative documents must be processed with reasonable accuracy, the detection and interpretation of tables is crucial. Table recognition has gained interest in document image analysis, in particular in unconstrained formats (absence of rule lines, unknown information of rows and columns). In this work, we propose a graph-based approach for detecting tables in document images. Instead of using the raw content (recognized text), we make use of the location, context and content type, thus it is purely a structure perception approach, not dependent on the language and the quality of the text
reading. Our framework makes use of Graph Neural Networks (GNNs) in order to describe the local repetitive structural information of tables in invoice documents. Our proposed model has been experimentally validated in two invoice datasets and achieved encouraging results. Additionally, due to the scarcity
of benchmark datasets for this task, we have contributed to the community a novel dataset derived from the RVL-CDIP invoice data. It will be publicly released to facilitate future research.
|
Kai Wang, Joost Van de Weijer, & Luis Herranz. (2021). ACAE-REMIND for online continual learning with compressed feature replay. PRL - Pattern Recognition Letters, 150, 122–129.
Abstract: Online continual learning aims to learn from a non-IID stream of data from a number of different tasks, where the learner is only allowed to consider data once. Methods are typically allowed to use a limited buffer to store some of the images in the stream. Recently, it was found that feature replay, where an intermediate layer representation of the image is stored (or generated) leads to superior results than image replay, while requiring less memory. Quantized exemplars can further reduce the memory usage. However, a drawback of these methods is that they use a fixed (or very intransigent) backbone network. This significantly limits the learning of representations that can discriminate between all tasks. To address this problem, we propose an auxiliary classifier auto-encoder (ACAE) module for feature replay at intermediate layers with high compression rates. The reduced memory footprint per image allows us to save more exemplars for replay. In our experiments, we conduct task-agnostic evaluation under online continual learning setting and get state-of-the-art performance on ImageNet-Subset, CIFAR100 and CIFAR10 dataset.
Keywords: online continual learning; autoencoders; vector quantization
|
Katerine Diaz, Konstantia Georgouli, Anastasios Koidis, & Jesus Martinez del Rincon. (2017). Incremental model learning for spectroscopy-based food analysis. CILS - Chemometrics and Intelligent Laboratory Systems, 167, 123–131.
Abstract: In this paper we propose the use of incremental learning for creating and improving multivariate analysis models in the field of chemometrics of spectral data. As main advantages, our proposed incremental subspace-based learning allows creating models faster, progressively improving previously created models and sharing them between laboratories and institutions without requiring transferring or disclosing individual spectra samples. In particular, our approach allows to improve the generalization and adaptability of previously generated models with a few new spectral samples to be applicable to real-world situations. The potential of our approach is demonstrated using vegetable oil type identification based on spectroscopic data as case study. Results show how incremental models maintain the accuracy of batch learning methodologies while reducing their computational cost and handicaps.
Keywords: Incremental model learning; IGDCV technique; Subspace based learning; IdentificationVegetable oils; FT-IR spectroscopy
|
Chris Bahnsen, David Vazquez, Antonio Lopez, & Thomas B. Moeslund. (2019). Learning to Remove Rain in Traffic Surveillance by Using Synthetic Data. In 14th International Conference on Computer Vision Theory and Applications (pp. 123–130).
Abstract: Rainfall is a problem in automated traffic surveillance. Rain streaks occlude the road users and degrade the overall visibility which in turn decrease object detection performance. One way of alleviating this is by artificially removing the rain from the images. This requires knowledge of corresponding rainy and rain-free images. Such images are often produced by overlaying synthetic rain on top of rain-free images. However, this method fails to incorporate the fact that rain fall in the entire three-dimensional volume of the scene. To overcome this, we introduce training data from the SYNTHIA virtual world that models rain streaks in the entirety of a scene. We train a conditional Generative Adversarial Network for rain removal and apply it on traffic surveillance images from SYNTHIA and the AAU RainSnow datasets. To measure the applicability of the rain-removed images in a traffic surveillance context, we run the YOLOv2 object detection algorithm on the original and rain-removed frames. The results on SYNTHIA show an 8% increase in detection accuracy compared to the original rain image. Interestingly, we find that high PSNR or SSIM scores do not imply good object detection performance.
Keywords: Rain Removal; Traffic Surveillance; Image Denoising
|
Jaume Garcia, Debora Gil, & Aura Hernandez-Sabate. (2010). Endowing Canonical Geometries to Cardiac Structures. In O. Camara, M. Pop, K. Rhode, M. Sermesant, N. Smith, & A. Young (Eds.), Statistical Atlases And Computational Models Of The Heart (Vol. 6364, pp. 124–133). LNCS. Springer Berlin / Heidelberg.
Abstract: International conference on Cardiac electrophysiological simulation challenge
In this paper, we show that canonical (shape-based) geometries can be endowed to cardiac structures using tubular coordinates defined over their medial axis. We give an analytic formulation of these geometries by means of B-Splines. Since B-Splines present vector space structure PCA can be applied to their control points and statistical models relating boundaries and the interior of the anatomical structures can be derived. We demonstrate the applicability in two cardiac structures, the 3D Left Ventricular volume, and the 2D Left-Right ventricle set in 2D Short Axis view.
|
Fadi Dornaika, Alireza Bosaghzadeh, & Bogdan Raducanu. (2013). Efficient Graph Construction for Label Propagation based Multi-observation Face Recognition. In Human Behavior Understanding 4th International Workshop (Vol. 8212, pp. 124–135). Springer International Publishing.
Abstract: Workshop on Human Behavior Understanding
Human-machine interaction is a hot topic nowadays in the communities of multimedia and computer vision. In this context, face recognition algorithms (used as primary cue for a person’s identity assessment) work well under controlled conditions but degrade significantly when tested in real-world environments. Recently, graph-based label propagation for multi-observation face recognition was proposed. However, the associated graphs were constructed in an ad-hoc manner (e.g., using the KNN graph) that cannot adapt optimally to the data. In this paper, we propose a novel approach for efficient and adaptive graph construction that can be used for multi-observation face recognition as well as for other recognition problems. Experimental results performed on Honda video face database, show a distinct advantage of the proposed method over the standard graph construction methods.
|
Patricia Marquez, Debora Gil, & Aura Hernandez-Sabate. (2012). A Complete Confidence Framework for Optical Flow. In Rita Cucchiara V. M. Andrea Fusiello (Ed.), 12th European Conference on Computer Vision – Workshops and Demonstrations (Vol. 7584, pp. 124–133). LNCS. Florence, Italy, October 7-13, 2012: Springer-Verlag.
Abstract: Medial representations are powerful tools for describing and parameterizing the volumetric shape of anatomical structures. Existing methods show excellent results when applied to 2D objects, but their quality drops across dimensions. This paper contributes to the computation of medial manifolds in two aspects. First, we provide a standard scheme for the computation of medial manifolds that avoid degenerated medial axis segments; second, we introduce an energy based method which performs independently of the dimension. We evaluate quantitatively the performance of our method with respect to existing approaches, by applying them to synthetic shapes of known medial geometry. Finally, we show results on shape representation of multiple abdominal organs, exploring the use of medial manifolds for the representation of multi-organ relations.
Keywords: Optical flow, confidence measures, sparsification plots, error prediction plots
|
Mariella Dimiccoli, Benoît Girard, Alain Berthoz, & Daniel Bennequin. (2013). Striola Magica: a functional explanation of otolith organs. JCN - Journal of Computational Neuroscience, 35(2), 125–154.
Abstract: Otolith end organs of vertebrates sense linear accelerations of the head and gravitation. The hair cells on their epithelia are responsible for transduction. In mammals, the striola, parallel to the line where hair cells reverse their polarization, is a narrow region centered on a curve with curvature and torsion. It has been shown that the striolar region is functionally different from the rest, being involved in a phasic vestibular pathway. We propose a mathematical and computational model that explains the necessity of this amazing geometry for the striola to be able to carry out its function. Our hypothesis, related to the biophysics of the hair cells and to the physiology of their afferent neurons, is that striolar afferents collect information from several type I hair cells to detect the jerk in a large domain of acceleration directions. This predicts a mean number of two calyces for afferent neurons, as measured in rodents. The domain of acceleration directions sensed by our striolar model is compatible with the experimental results obtained on monkeys considering all afferents. Therefore, the main result of our study is that phasic and tonic vestibular afferents cover the same geometrical fields, but at different dynamical and frequency domains.
Keywords: Otolith organs ;Striola; Vestibular pathway
|
Idoia Ruiz, Lorenzo Porzi, Samuel Rota Bulo, Peter Kontschieder, & Joan Serrat. (2021). Weakly Supervised Multi-Object Tracking and Segmentation. In IEEE Winter Conference on Applications of Computer Vision Workshops (pp. 125–133).
Abstract: We introduce the problem of weakly supervised MultiObject Tracking and Segmentation, i.e. joint weakly supervised instance segmentation and multi-object tracking, in which we do not provide any kind of mask annotation.
To address it, we design a novel synergistic training strategy by taking advantage of multi-task learning, i.e. classification and tracking tasks guide the training of the unsupervised instance segmentation. For that purpose, we extract weak foreground localization information, provided by
Grad-CAM heatmaps, to generate a partial ground truth to learn from. Additionally, RGB image level information is employed to refine the mask prediction at the edges of the
objects. We evaluate our method on KITTI MOTS, the most representative benchmark for this task, reducing the performance gap on the MOTSP metric between the fully supervised and weakly supervised approach to just 12% and 12.7 % for cars and pedestrians, respectively.
|
Marina Alberti, Carlo Gatta, Simone Balocco, Francesco Ciompi, Oriol Pujol, Joana Silva, et al. (2011). Automatic Branching Detection in IVUS Sequences. In Jordi Vitria, Joao Miguel Raposo, & Mario Hernandez (Eds.), 5th Iberian Conference on Pattern Recognition and Image Analysis (Vol. 6669, pp. 126–133). LNCS. Berlin: Springer Berlin Heidelberg.
Abstract: Atherosclerosis is a vascular pathology affecting the arterial walls, generally located in specific vessel sites, such as bifurcations. In this paper, for the first time, a fully automatic approach for the detection of bifurcations in IVUS pullback sequences is presented. The method identifies the frames and the angular sectors in which a bifurcation is visible. This goal is achieved by applying a classifier to a set of textural features extracted from each image of an IVUS pullback. A comparison between two state-of-the-art classifiers is performed, AdaBoost and Random Forest. A cross-validation scheme is applied in order to evaluate the performances of the approaches. The obtained results are encouraging, showing a sensitivity of 75% and an accuracy of 94% by using the AdaBoost algorithm.
|
Miguel Angel Bautista, Antonio Hernandez, Victor Ponce, Xavier Perez Sala, Xavier Baro, Oriol Pujol, et al. (2012). Probability-based Dynamic TimeWarping for Gesture Recognition on RGB-D data. In 21st International Conference on Pattern Recognition International Workshop on Depth Image Analysis (Vol. 7854, pp. 126–135). Springer Berlin Heidelberg.
Abstract: Dynamic Time Warping (DTW) is commonly used in gesture recognition tasks in order to tackle the temporal length variability of gestures. In the DTW framework, a set of gesture patterns are compared one by one to a maybe infinite test sequence, and a query gesture category is recognized if a warping cost below a certain threshold is found within the test sequence. Nevertheless, either taking one single sample per gesture category or a set of isolated samples may not encode the variability of such gesture category. In this paper, a probability-based DTW for gesture recognition is proposed. Different samples of the same gesture pattern obtained from RGB-Depth data are used to build a Gaussian-based probabilistic model of the gesture. Finally, the cost of DTW has been adapted accordingly to the new model. The proposed approach is tested in a challenging scenario, showing better performance of the probability-based DTW in comparison to state-of-the-art approaches for gesture recognition on RGB-D data.
|
Arjan Gijsenij, Theo Gevers, & Joost Van de Weijer. (2010). Generalized Gamut Mapping using Image Derivative Structures for Color Constancy. IJCV - International Journal of Computer Vision, 86(2-3), 127–139.
Abstract: The gamut mapping algorithm is one of the most promising methods to achieve computational color constancy. However, so far, gamut mapping algorithms are restricted to the use of pixel values to estimate the illuminant. Therefore, in this paper, gamut mapping is extended to incorporate the statistical nature of images. It is analytically shown that the proposed gamut mapping framework is able to include any linear filter output. The main focus is on the local n-jet describing the derivative structure of an image. It is shown that derivatives have the advantage over pixel values to be invariant to disturbing effects (i.e. deviations of the diagonal model) such as saturated colors and diffuse light. Further, as the n-jet based gamut mapping has the ability to use more information than pixel values alone, the combination of these algorithms are more stable than the regular gamut mapping algorithm. Different methods of combining are proposed. Based on theoretical and experimental results conducted on large scale data sets of hyperspectral, laboratory and realworld scenes, it can be derived that (1) in case of deviations of the diagonal model, the derivative-based approach outperforms the pixel-based gamut mapping, (2) state-of-the-art algorithms are outperformed by the n-jet based gamut mapping, (3) the combination of the different n-jet based gamut
|
|
Joana Maria Pujadas-Mora, Alicia Fornes, Josep Llados, & Anna Cabre. (2016). Bridging the gap between historical demography and computing: tools for computer-assisted transcription and the analysis of demographic sources. In K.Matthijs, S.Hin, H.Matsuo, & J.Kok (Eds.), The future of historical demography. Upside down and inside out (pp. 127–131). Acco Publishers.
|
B. Gautam, Oriol Ramos Terrades, Joana Maria Pujadas-Mora, & Miquel Valls-Figols. (2020). Knowledge graph based methods for record linkage. PRL - Pattern Recognition Letters, 136, 127–133.
Abstract: Nowadays, it is common in Historical Demography the use of individual-level data as a consequence of a predominant life-course approach for the understanding of the demographic behaviour, family transition, mobility, etc. Advanced record linkage is key since it allows increasing the data complexity and its volume to be analyzed. However, current methods are constrained to link data from the same kind of sources. Knowledge graph are flexible semantic representations, which allow to encode data variability and semantic relations in a structured manner.
In this paper we propose the use of knowledge graph methods to tackle record linkage tasks. The proposed method, named WERL, takes advantage of the main knowledge graph properties and learns embedding vectors to encode census information. These embeddings are properly weighted to maximize the record linkage performance. We have evaluated this method on benchmark data sets and we have compared it to related methods with stimulating and satisfactory results.
|
Razieh Rastgoo, Kourosh Kiani, & Sergio Escalera. (2020). Hand pose aware multimodal isolated sign language recognition. MTAP - Multimedia Tools and Applications, 80, 127–163.
Abstract: Isolated hand sign language recognition from video is a challenging research area in computer vision. Some of the most important challenges in this area include dealing with hand occlusion, fast hand movement, illumination changes, or background complexity. While most of the state-of-the-art results in the field have been achieved using deep learning-based models, the previous challenges are not completely solved. In this paper, we propose a hand pose aware model for isolated hand sign language recognition using deep learning approaches from two input modalities, RGB and depth videos. Four spatial feature types: pixel-level, flow, deep hand, and hand pose features, fused from both visual modalities, are input to LSTM for temporal sign recognition. While we use Optical Flow (OF) for flow information in RGB video inputs, Scene Flow (SF) is used for depth video inputs. By including hand pose features, we show a consistent performance improvement of the sign language recognition model. To the best of our knowledge, this is the first time that this discriminant spatiotemporal features, benefiting from the hand pose estimation features and multi-modal inputs, are fused for isolated hand sign language recognition. We perform a step-by-step analysis of the impact in terms of recognition performance of the hand pose features, different combinations of the spatial features, and different recurrent models, especially LSTM and GRU. Results on four public datasets confirm that the proposed model outperforms the current state-of-the-art models on Montalbano II, MSR Daily Activity 3D, and CAD-60 datasets with a relative accuracy improvement of 1.64%, 6.5%, and 7.6%. Furthermore, our model obtains a competitive results on isoGD dataset with only 0.22% margin lower than the current state-of-the-art model.
|
