Y. Patel, Lluis Gomez, Raul Gomez, Marçal Rusiñol, Dimosthenis Karatzas, & C.V. Jawahar. (2018). TextTopicNet-Self-Supervised Learning of Visual Features Through Embedding Images on Semantic Text Spaces.
Abstract: The immense success of deep learning based methods in computer vision heavily relies on large scale training datasets. These richly annotated datasets help the network learn discriminative visual features. Collecting and annotating such datasets requires a tremendous amount of human effort and annotations are limited to popular set of classes. As an alternative, learning visual features by designing auxiliary tasks which make use of freely available self-supervision has become increasingly popular in the computer vision community.
In this paper, we put forward an idea to take advantage of multi-modal context to provide self-supervision for the training of computer vision algorithms. We show that adequate visual features can be learned efficiently by training a CNN to predict the semantic textual context in which a particular image is more probable to appear as an illustration. More specifically we use popular text embedding techniques to provide the self-supervision for the training of deep CNN.
|
Anguelos Nicolaou, Sounak Dey, V.Christlein, A.Maier, & Dimosthenis Karatzas. (2018). Non-deterministic Behavior of Ranking-based Metrics when Evaluating Embeddings. In International Workshop on Reproducible Research in Pattern Recognition (Vol. 11455, pp. 71–82). LNCS.
Abstract: Embedding data into vector spaces is a very popular strategy of pattern recognition methods. When distances between embeddings are quantized, performance metrics become ambiguous. In this paper, we present an analysis of the ambiguity quantized distances introduce and provide bounds on the effect. We demonstrate that it can have a measurable effect in empirical data in state-of-the-art systems. We also approach the phenomenon from a computer security perspective and demonstrate how someone being evaluated by a third party can exploit this ambiguity and greatly outperform a random predictor without even access to the input data. We also suggest a simple solution making the performance metrics, which rely on ranking, totally deterministic and impervious to such exploits.
|
Dena Bazazian, Dimosthenis Karatzas, & Andrew Bagdanov. (2018). Word Spotting in Scene Images based on Character Recognition. In IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (pp. 1872–1874).
Abstract: In this paper we address the problem of unconstrained Word Spotting in scene images. We train a Fully Convolutional Network to produce heatmaps of all the character classes. Then, we employ the Text Proposals approach and, via a rectangle classifier, detect the most likely rectangle for each query word based on the character attribute maps. We evaluate the proposed method on ICDAR2015 and show that it is capable of identifying and recognizing query words in natural scene images.
|
Zhijie Fang, & Antonio Lopez. (2018). Is the Pedestrian going to Cross? Answering by 2D Pose Estimation. In IEEE Intelligent Vehicles Symposium (pp. 1271–1276).
Abstract: Our recent work suggests that, thanks to nowadays powerful CNNs, image-based 2D pose estimation is a promising cue for determining pedestrian intentions such as crossing the road in the path of the ego-vehicle, stopping before entering the road, and starting to walk or bending towards the road. This statement is based on the results obtained on non-naturalistic sequences (Daimler dataset), i.e. in sequences choreographed specifically for performing the study. Fortunately, a new publicly available dataset (JAAD) has appeared recently to allow developing methods for detecting pedestrian intentions in naturalistic driving conditions; more specifically, for addressing the relevant question is the pedestrian going to cross? Accordingly, in this paper we use JAAD to assess the usefulness of 2D pose estimation for answering such a question. We combine CNN-based pedestrian detection, tracking and pose estimation to predict the crossing action from monocular images. Overall, the proposed pipeline provides new state-ofthe-art results.
|
Jiaolong Xu, Peng Wang, Heng Yang, & Antonio Lopez. (2019). Training a Binary Weight Object Detector by Knowledge Transfer for Autonomous Driving. In IEEE International Conference on Robotics and Automation (pp. 2379–2384).
Abstract: Autonomous driving has harsh requirements of small model size and energy efficiency, in order to enable the embedded system to achieve real-time on-board object detection. Recent deep convolutional neural network based object detectors have achieved state-of-the-art accuracy. However, such models are trained with numerous parameters and their high computational costs and large storage prohibit the deployment to memory and computation resource limited systems. Low-precision neural networks are popular techniques for reducing the computation requirements and memory footprint. Among them, binary weight neural network (BWN) is the extreme case which quantizes the float-point into just bit. BWNs are difficult to train and suffer from accuracy deprecation due to the extreme low-bit representation. To address this problem, we propose a knowledge transfer (KT) method to aid the training of BWN using a full-precision teacher network. We built DarkNet-and MobileNet-based binary weight YOLO-v2 detectors and conduct experiments on KITTI benchmark for car, pedestrian and cyclist detection. The experimental results show that the proposed method maintains high detection accuracy while reducing the model size of DarkNet-YOLO from 257 MB to 8.8 MB and MobileNet-YOLO from 193 MB to 7.9 MB.
|
Akhil Gurram, Onay Urfalioglu, Ibrahim Halfaoui, Fahd Bouzaraa, & Antonio Lopez. (2018). Monocular Depth Estimation by Learning from Heterogeneous Datasets. In IEEE Intelligent Vehicles Symposium (pp. 2176–2181).
Abstract: Depth estimation provides essential information to perform autonomous driving and driver assistance. Especially, Monocular Depth Estimation is interesting from a practical point of view, since using a single camera is cheaper than many other options and avoids the need for continuous calibration strategies as required by stereo-vision approaches. State-of-the-art methods for Monocular Depth Estimation are based on Convolutional Neural Networks (CNNs). A promising line of work consists of introducing additional semantic information about the traffic scene when training CNNs for depth estimation. In practice, this means that the depth data used for CNN training is complemented with images having pixel-wise semantic labels, which usually are difficult to annotate (eg crowded urban images). Moreover, so far it is common practice to assume that the same raw training data is associated with both types of ground truth, ie, depth and semantic labels. The main contribution of this paper is to show that this hard constraint can be circumvented, ie, that we can train CNNs for depth estimation by leveraging the depth and semantic information coming from heterogeneous datasets. In order to illustrate the benefits of our approach, we combine KITTI depth and Cityscapes semantic segmentation datasets, outperforming state-of-the-art results on Monocular Depth Estimation.
|
Alejandro Cartas, Estefania Talavera, Petia Radeva, & Mariella Dimiccoli. (2018). On the Role of Event Boundaries in Egocentric Activity Recognition from Photostreams.
Abstract: Event boundaries play a crucial role as a pre-processing step for detection, localization, and recognition tasks of human activities in videos. Typically, although their intrinsic subjectiveness, temporal bounds are provided manually as input for training action recognition algorithms. However, their role for activity recognition in the domain of egocentric photostreams has been so far neglected. In this paper, we provide insights of how automatically computed boundaries can impact activity recognition results in the emerging domain of egocentric photostreams. Furthermore, we collected a new annotated dataset acquired by 15 people by a wearable photo-camera and we used it to show the generalization capabilities of several deep learning based architectures to unseen users.
|
Md. Mostafa Kamal Sarker, Hatem A. Rashwan, Hatem A. Rashwan, Estefania Talavera, Syeda Furruka Banu, Petia Radeva, et al. (2018). MACNet: Multi-scale Atrous Convolution Networks for Food Places Classification in Egocentric Photo-streams. In European Conference on Computer Vision workshops (pp. 423–433). LCNS.
Abstract: First-person (wearable) camera continually captures unscripted interactions of the camera user with objects, people, and scenes reflecting his personal and relational tendencies. One of the preferences of people is their interaction with food events. The regulation of food intake and its duration has a great importance to protect against diseases. Consequently, this work aims to develop a smart model that is able to determine the recurrences of a person on food places during a day. This model is based on a deep end-to-end model for automatic food places recognition by analyzing egocentric photo-streams. In this paper, we apply multi-scale Atrous convolution networks to extract the key features related to food places of the input images. The proposed model is evaluated on an in-house private dataset called “EgoFoodPlaces”. Experimental results shows promising results of food places classification recognition in egocentric photo-streams.
|
Marçal Rusiñol, Dimosthenis Karatzas, & Josep Llados. (2015). Automatic Verification of Properly Signed Multi-page Document Images. In Proceedings of the Eleventh International Symposium on Visual Computing (Vol. 9475, pp. 327–336). LNCS, 9475.
Abstract: In this paper we present an industrial application for the automatic screening of incoming multi-page documents in a banking workflow aimed at determining whether these documents are properly signed or not. The proposed method is divided in three main steps. First individual pages are classified in order to identify the pages that should contain a signature. In a second step, we segment within those key pages the location where the signatures should appear. The last step checks whether the signatures are present or not. Our method is tested in a real large-scale environment and we report the results when checking two different types of real multi-page contracts, having in total more than 14,500 pages.
Keywords: Document Image; Manual Inspection; Signature Verification; Rejection Criterion; Document Flow
|
L. Rothacker, Marçal Rusiñol, Josep Llados, & G.A. Fink. (2014). A Two-stage Approach to Segmentation-Free Query-by-example Word Spotting. Manuscript Cultures, 47–58.
Abstract: With the ongoing progress in digitization, huge document collections and archives have become available to a broad audience. Scanned document images can be transmitted electronically and studied simultaneously throughout the world. While this is very beneficial, it is often impossible to perform automated searches on these document collections. Optical character recognition usually fails when it comes to handwritten or historic documents. In order to address the need for exploring document collections rapidly, researchers are working on word spotting. In query-by-example word spotting scenarios, the user selects an exemplary occurrence of the query word in a document image. The word spotting system then retrieves all regions in the collection that are visually similar to the given example of the query word. The best matching regions are presented to the user and no actual transcription is required.
An important property of a word spotting system is the computational speed with which queries can be executed. In our previous work, we presented a relatively slow but high-precision method. In the present work, we will extend this baseline system to an integrated two-stage approach. In a coarse-grained first stage, we will filter document images efficiently in order to identify regions that are likely to contain the query word. In the fine-grained second stage, these regions will be analyzed with our previously presented high-precision method. Finally, we will report recognition results and query times for the well-known George Washington
benchmark in our evaluation. We achieve state-of-the-art recognition results while the query times can be reduced to 50% in comparison with our baseline.
|
Xavier Soria, & Angel Sappa. (2018). Improving Edge Detection in RGB Images by Adding NIR Channel. In 14th IEEE International Conference on Signal Image Technology & Internet Based System.
Abstract: The edge detection is yet a critical problem in many computer vision and image processing tasks. The manuscript presents an Holistically-Nested Edge Detection based approach to study the inclusion of Near-Infrared in the Visible spectrum
images. To do so, a Single Sensor based dataset has been acquired in the range of 400nm to 1100nm wavelength spectral band. Prominent results have been obtained even when the ground truth (annotated edge-map) is based in the visible wavelength spectrum.
Keywords: Edge detection; Contour detection; VGG; CNN; RGB-NIR; Near infrared images
|
Patricia Suarez, Angel Sappa, & Boris X. Vintimilla. (2018). Cross-spectral image dehaze through a dense stacked conditional GAN based approach. In 14th IEEE International Conference on Signal Image Technology & Internet Based System.
Abstract: This paper proposes a novel approach to remove haze from RGB images using a near infrared images based on a dense stacked conditional Generative Adversarial Network (CGAN). The architecture of the deep network implemented
receives, besides the images with haze, its corresponding image in the near infrared spectrum, which serve to accelerate the learning process of the details of the characteristics of the images. The model uses a triplet layer that allows the independence learning of each channel of the visible spectrum image to remove the haze on each color channel separately. A multiple loss function scheme is proposed, which ensures balanced learning between the colors
and the structure of the images. Experimental results have shown that the proposed method effectively removes the haze from the images. Additionally, the proposed approach is compared with a state of the art approach showing better results.
Keywords: Infrared imaging; Dense; Stacked CGAN; Crossspectral; Convolutional networks
|
Jorge Charco, Boris X. Vintimilla, & Angel Sappa. (2018). Deep learning based camera pose estimation in multi-view environment. In 14th IEEE International Conference on Signal Image Technology & Internet Based System.
Abstract: This paper proposes to use a deep learning network architecture for relative camera pose estimation on a multi-view environment. The proposed network is a variant architecture of AlexNet to use as regressor for prediction the relative translation and rotation as output. The proposed approach is trained from
scratch on a large data set that takes as input a pair of imagesfrom the same scene. This new architecture is compared with a previous approach using standard metrics, obtaining better results on the relative camera pose.
Keywords: Deep learning; Camera pose estimation; Multiview environment; Siamese architecture
|
Patricia Suarez, Angel Sappa, Boris X. Vintimilla, & Riad I. Hammoud. (2018). Near InfraRed Imagery Colorization. In 25th International Conference on Image Processing (pp. 2237–2241).
Abstract: This paper proposes a stacked conditional Generative Adversarial Network-based method for Near InfraRed (NIR) imagery colorization. We propose a variant architecture of Generative Adversarial Network (GAN) that uses multiple
loss functions over a conditional probabilistic generative model. We show that this new architecture/loss-function yields better generalization and representation of the generated colored IR images. The proposed approach is evaluated on a large test dataset and compared to recent state of the art methods using standard metrics.
Keywords: Convolutional Neural Networks (CNN), Generative Adversarial Network (GAN), Infrared Imagery colorization
|
Patricia Suarez, Angel Sappa, & Boris X. Vintimilla. (2018). Vegetation Index Estimation from Monospectral Images. In 15th International Conference on Images Analysis and Recognition (Vol. 10882, pp. 353–362). LNCS.
Abstract: This paper proposes a novel approach to estimate Normalized Difference Vegetation Index (NDVI) from just the red channel of a RGB image. The NDVI index is defined as the ratio of the difference of the red and infrared radiances over their sum. In other words, information from the red channel of a RGB image and the corresponding infrared spectral band are required for its computation. In the current work the NDVI index is estimated just from the red channel by training a Conditional Generative Adversarial Network (CGAN). The architecture proposed for the generative network consists of a single level structure, which combines at the final layer results from convolutional operations together with the given red channel with Gaussian noise to enhance
details, resulting in a sharp NDVI image. Then, the discriminative model
estimates the probability that the NDVI generated index came from the training dataset, rather than the index automatically generated. Experimental results with a large set of real images are provided showing that a Conditional GAN single level model represents an acceptable approach to estimate NDVI index.
|
