Hazim Kemal Ekenel

Slavisa Aleksic, Michael Atanasov, Jean Calleja Agius et al.

The report illustrates the state of the art of the most successful AAL applications and functions based on audio and video data, namely (i) lifelogging and self-monitoring, (ii) remote monitoring of vital signs, (iii) emotional state recognition, (iv) food intake monitoring, activity and behaviour recognition, (v) activity and personal assistance, (vi) gesture recognition, (vii) fall detection and prevention, (viii) mobility assessment and frailty recognition, and (ix) cognitive and motor rehabilitation. For these application scenarios, the report illustrates the state of play in terms of scientific advances, available products and research project. The open challenges are also highlighted.

Dogucan Yaman, Fevziye Irem Eyiokur, Leonard Bärmann et al.

Talking face generation aims to create realistic videos with accurate lip synchronization and high visual quality, using given audio and reference video while preserving identity and visual characteristics. In this paper, we start by identifying several issues with existing synchronization learning methods. These involve unstable training, lip synchronization, and visual quality issues caused by lip-sync loss, SyncNet, and lip leaking from the identity reference. To address these issues, we first tackle the lip leaking problem by introducing a silent-lip generator, which changes the lips of the identity reference to alleviate leakage. We then introduce stabilized synchronization loss and AVSyncNet to overcome problems caused by lip-sync loss and SyncNet. Experiments show that our model outperforms state-of-the-art methods in both visual quality and lip synchronization. Comprehensive ablation studies further validate our individual contributions and their cohesive effects.

Fabio Valerio Massoli, Fabrizio Falchi, Alperen Kantarci et al.

Anomalies are ubiquitous in all scientific fields and can express an unexpected event due to incomplete knowledge about the data distribution or an unknown process that suddenly comes into play and distorts observations. Due to such events' rarity, to train deep learning models on the Anomaly Detection (AD) task, scientists only rely on "normal" data, i.e., non-anomalous samples. Thus, letting the neural network infer the distribution beneath the input data. In such a context, we propose a novel framework, named Multi-layer One-Class ClassificAtion (MOCCA),to train and test deep learning models on the AD task. Specifically, we applied it to autoencoders. A key novelty in our work stems from the explicit optimization of intermediate representations for the AD task. Indeed, differently from commonly used approaches that consider a neural network as a single computational block, i.e., using the output of the last layer only, MOCCA explicitly leverages the multi-layer structure of deep architectures. Each layer's feature space is optimized for AD during training, while in the test phase, the deep representations extracted from the trained layers are combined to detect anomalies. With MOCCA, we split the training process into two steps. First, the autoencoder is trained on the reconstruction task only. Then, we only retain the encoder tasked with minimizing the L_2 distance between the output representation and a reference point, the anomaly-free training data centroid, at each considered layer. Subsequently, we combine the deep features extracted at the various trained layers of the encoder model to detect anomalies at inference time. To assess the performance of the models trained with MOCCA, we conduct extensive experiments on publicly available datasets. We show that our proposed method reaches comparable or superior performance to state-of-the-art approaches available in the literature.

Mohammad Saeed Rad, Thomas Yu, Claudiu Musat et al.

Super-resolution (SR) has traditionally been based on pairs of high-resolution images (HR) and their low-resolution (LR) counterparts obtained artificially with bicubic downsampling. However, in real-world SR, there is a large variety of realistic image degradations and analytically modeling these realistic degradations can prove quite difficult. In this work, we propose to handle real-world SR by splitting this ill-posed problem into two comparatively more well-posed steps. First, we train a network to transform real LR images to the space of bicubically downsampled images in a supervised manner, by using both real LR/HR pairs and synthetic pairs. Second, we take a generic SR network trained on bicubically downsampled images to super-resolve the transformed LR image. The first step of the pipeline addresses the problem by registering the large variety of degraded images to a common, well understood space of images. The second step then leverages the already impressive performance of SR on bicubically downsampled images, sidestepping the issues of end-to-end training on datasets with many different image degradations. We demonstrate the effectiveness of our proposed method by comparing it to recent methods in real-world SR and show that our proposed approach outperforms the state-of-the-art works in terms of both qualitative and quantitative results, as well as results of an extensive user study conducted on several real image datasets.

Mohammad Saeed Rad, Behzad Bozorgtabar, Urs-Viktor Marti et al.

By benefiting from perceptual losses, recent studies have improved significantly the performance of the super-resolution task, where a high-resolution image is resolved from its low-resolution counterpart. Although such objective functions generate near-photorealistic results, their capability is limited, since they estimate the reconstruction error for an entire image in the same way, without considering any semantic information. In this paper, we propose a novel method to benefit from perceptual loss in a more objective way. We optimize a deep network-based decoder with a targeted objective function that penalizes images at different semantic levels using the corresponding terms. In particular, the proposed method leverages our proposed OBB (Object, Background and Boundary) labels, generated from segmentation labels, to estimate a suitable perceptual loss for boundaries, while considering texture similarity for backgrounds. We show that our proposed approach results in more realistic textures and sharper edges, and outperforms other state-of-the-art algorithms in terms of both qualitative results on standard benchmarks and results of extensive user studies.

Mohammad Saeed Rad, Behzad Bozorgtabar, Claudiu Musat et al.

Despite significant progress toward super resolving more realistic images by deeper convolutional neural networks (CNNs), reconstructing fine and natural textures still remains a challenging problem. Recent works on single image super resolution (SISR) are mostly based on optimizing pixel and content wise similarity between recovered and high-resolution (HR) images and do not benefit from recognizability of semantic classes. In this paper, we introduce a novel approach using categorical information to tackle the SISR problem; we present a decoder architecture able to extract and use semantic information to super-resolve a given image by using multitask learning, simultaneously for image super-resolution and semantic segmentation. To explore categorical information during training, the proposed decoder only employs one shared deep network for two task-specific output layers. At run-time only layers resulting HR image are used and no segmentation label is required. Extensive perceptual experiments and a user study on images randomly selected from COCO-Stuff dataset demonstrate the effectiveness of our proposed method and it outperforms the state-of-the-art methods.

Guillaume Jaume, Hazim Kemal Ekenel, Jean-Philippe Thiran

We present a new dataset for form understanding in noisy scanned documents (FUNSD) that aims at extracting and structuring the textual content of forms. The dataset comprises 199 real, fully annotated, scanned forms. The documents are noisy and vary widely in appearance, making form understanding (FoUn) a challenging task. The proposed dataset can be used for various tasks, including text detection, optical character recognition, spatial layout analysis, and entity labeling/linking. To the best of our knowledge, this is the first publicly available dataset with comprehensive annotations to address FoUn task. We also present a set of baselines and introduce metrics to evaluate performance on the FUNSD dataset, which can be downloaded at https://guillaumejaume.github.io/FUNSD/.

Behzad Bozorgtabar, Mohammad Saeed Rad, Hazim Kemal Ekenel et al.

Cross-domain synthesizing realistic faces to learn deep models has attracted increasing attention for facial expression analysis as it helps to improve the performance of expression recognition accuracy despite having small number of real training images. However, learning from synthetic face images can be problematic due to the distribution discrepancy between low-quality synthetic images and real face images and may not achieve the desired performance when the learned model applies to real world scenarios. To this end, we propose a new attribute guided face image synthesis to perform a translation between multiple image domains using a single model. In addition, we adopt the proposed model to learn from synthetic faces by matching the feature distributions between different domains while preserving each domain's characteristics. We evaluate the effectiveness of the proposed approach on several face datasets on generating realistic face images. We demonstrate that the expression recognition performance can be enhanced by benefiting from our face synthesis model. Moreover, we also conduct experiments on a near-infrared dataset containing facial expression videos of drivers to assess the performance using in-the-wild data for driver emotion recognition.

Guillaume Jaume, Behzad Bozorgtabar, Hazim Kemal Ekenel et al.

We introduce a new scene graph generation method called image-level attentional context modeling (ILAC). Our model includes an attentional graph network that effectively propagates contextual information across the graph using image-level features. Whereas previous works use an object-centric context, we build an image-level context agent to encode the scene properties. The proposed method comprises a single-stream network that iteratively refines the scene graph with a nested graph neural network. We demonstrate that our approach achieves competitive performance with the state-of-the-art for scene graph generation on the Visual Genome dataset, while requiring fewer parameters than other methods. We also show that ILAC can improve regular object detectors by incorporating relational image-level information.

Mohammad Saeed Rad, Andreas von Kaenel, Andre Droux et al.

Littering quantification is an important step for improving cleanliness of cities. When human interpretation is too cumbersome or in some cases impossible, an objective index of cleanliness could reduce the littering by awareness actions. In this paper, we present a fully automated computer vision application for littering quantification based on images taken from the streets and sidewalks. We have employed a deep learning based framework to localize and classify different types of wastes. Since there was no waste dataset available, we built our acquisition system mounted on a vehicle. Collected images containing different types of wastes. These images are then annotated for training and benchmarking the developed system. Our results on real case scenarios show accurate detection of littering on variant backgrounds.

Klemen Grm, Vitomir Štruc, Anais Artiges et al.

Deep convolutional neural networks (CNNs) based approaches are the state-of-the-art in various computer vision tasks, including face recognition. Considerable research effort is currently being directed towards further improving deep CNNs by focusing on more powerful model architectures and better learning techniques. However, studies systematically exploring the strengths and weaknesses of existing deep models for face recognition are still relatively scarce in the literature. In this paper, we try to fill this gap and study the effects of different covariates on the verification performance of four recent deep CNN models using the Labeled Faces in the Wild (LFW) dataset. Specifically, we investigate the influence of covariates related to: image quality -- blur, JPEG compression, occlusion, noise, image brightness, contrast, missing pixels; and model characteristics -- CNN architecture, color information, descriptor computation; and analyze their impact on the face verification performance of AlexNet, VGG-Face, GoogLeNet, and SqueezeNet. Based on comprehensive and rigorous experimentation, we identify the strengths and weaknesses of the deep learning models, and present key areas for potential future research. Our results indicate that high levels of noise, blur, missing pixels, and brightness have a detrimental effect on the verification performance of all models, whereas the impact of contrast changes and compression artifacts is limited. It has been found that the descriptor computation strategy and color information does not have a significant influence on performance.

Samil Karahan, Merve Kilinc Yildirim, Kadir Kirtac et al.

Face recognition approaches that are based on deep convolutional neural networks (CNN) have been dominating the field. The performance improvements they have provided in the so called in-the-wild datasets are significant, however, their performance under image quality degradations have not been assessed, yet. This is particularly important, since in real-world face recognition applications, images may contain various kinds of degradations due to motion blur, noise, compression artifacts, color distortions, and occlusion. In this work, we have addressed this problem and analyzed the influence of these image degradations on the performance of deep CNN-based face recognition approaches using the standard LFW closed-set identification protocol. We have evaluated three popular deep CNN models, namely, the AlexNet, VGG-Face, and GoogLeNet. Results have indicated that blur, noise, and occlusion cause a significant decrease in performance, while deep CNN models are found to be robust to distortions, such as color distortions and change in color balance.

Refik Can Malli, Mehmet Aygun, Hazim Kemal Ekenel

In this paper, we address the problem of apparent age estimation. Different from estimating the real age of individuals, in which each face image has a single age label, in this problem, face images have multiple age labels, corresponding to the ages perceived by the annotators, when they look at these images. This provides an intriguing computer vision problem, since in generic image or object classification tasks, it is typical to have a single ground truth label per class. To account for multiple labels per image, instead of using average age of the annotated face image as the class label, we have grouped the face images that are within a specified age range. Using these age groups and their age-shifted groupings, we have trained an ensemble of deep learning models. Before feeding an input face image to a deep learning model, five facial landmark points are detected and used for 2-D alignment. We have employed and fine tuned convolutional neural networks (CNNs) that are based on VGG-16 [24] architecture and pretrained on the IMDB-WIKI dataset [22]. The outputs of these deep learning models are then combined to produce the final estimation. Proposed method achieves 0.3668 error in the final ChaLearn LAP 2016 challenge test set [5].

Mostafa Mehdipour Ghazi, Hazim Kemal Ekenel

Deep learning based approaches have been dominating the face recognition field due to the significant performance improvement they have provided on the challenging wild datasets. These approaches have been extensively tested on such unconstrained datasets, on the Labeled Faces in the Wild and YouTube Faces, to name a few. However, their capability to handle individual appearance variations caused by factors such as head pose, illumination, occlusion, and misalignment has not been thoroughly assessed till now. In this paper, we present a comprehensive study to evaluate the performance of deep learning based face representation under several conditions including the varying head pose angles, upper and lower face occlusion, changing illumination of different strengths, and misalignment due to erroneous facial feature localization. Two successful and publicly available deep learning models, namely VGG-Face and Lightened CNN have been utilized to extract face representations. The obtained results show that although deep learning provides a powerful representation for face recognition, it can still benefit from preprocessing, for example, for pose and illumination normalization in order to achieve better performance under various conditions. Particularly, if these variations are not included in the dataset used to train the deep learning model, the role of preprocessing becomes more crucial. Experimental results also show that deep learning based representation is robust to misalignment and can tolerate facial feature localization errors up to 10% of the interocular distance.