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.

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.

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.

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.

Abstract: This paper proposes a novel approach to remove haze degradations in RGB images using a stacked conditional Generative Adversarial Network (GAN). It employs a triplet of GAN to remove the haze on each color channel independently.
A multiple loss functions scheme, applied over a conditional probabilistic model, is proposed. The proposed GAN architecture learns to remove the haze, using as conditioned entrance, the images with haze from which the clear
images will be obtained. Such formulation ensures a fast model training convergence and a homogeneous model generalization. Experiments showed that the proposed method generates high-quality clear images.

Abstract: Future video prediction is an ill-posed Computer Vision problem that recently received much attention. Its main challenges are the high variability in video content, the propagation of errors through time, and the non-specificity of the future frames: given a sequence of past frames there is a continuous distribution of possible futures. This work introduces bijective Gated Recurrent Units, a double mapping between the input and output of a GRU layer. This allows for recurrent auto-encoders with state sharing between encoder and decoder, stratifying the sequence representation and helping to prevent capacity problems. We show how with this topology only the encoder or decoder needs to be applied for input encoding and prediction, respectively. This reduces the computational cost and avoids re-encoding the predictions when generating a sequence of frames, mitigating the propagation of errors. Furthermore, it is possible to remove layers from an already trained model, giving an insight to the role performed by each layer and making the model more explainable. We evaluate our approach on three video datasets, outperforming state of the art prediction results on MMNIST and UCF101, and obtaining competitive results on KTH with 2 and 3 times less memory usage and computational cost than the best scored approach.

Abstract: Action recognition pipelines that use Recurrent Neural Networks (RNN) are currently 5–10% less accurate than Convolutional Neural Networks (CNN). While most works that use RNNs employ a 2D CNN on each frame to extract descriptors for action recognition, we extract spatiotemporal features from a 3D CNN and then learn the temporal relationship of these descriptors through a stacked residual recurrent neural network (Res-RNN). We introduce for the first time residual learning to counter the degradation problem in multi-layer RNNs, which have been successful for temporal aggregation in two-stream action recognition pipelines. Finally, we use a late fusion strategy to combine RGB and optical flow data of the two-stream Res-RNN. Experimental results show that the proposed pipeline achieves competitive results on UCF-101 and state of-the-art results for RNN-like architectures on the challenging HMDB-51 dataset.