Abstract: On-board vision systems may need to increase the number of classes that can be recognized in a relatively short period. For instance, a traffic sign recognition system may suddenly be required to recognize new signs. Since collecting and annotating samples of such new classes may need more time than we wish, especially for uncommon signs, we propose a method to generate these samples by combining synthetic images and Generative Adversarial Network (GAN) technology. In particular, the GAN is trained on synthetic and real-world samples from known classes to perform synthetic-to-real domain adaptation, but applied to synthetic samples of the new classes. Using the Tsinghua dataset with a synthetic counterpart, SYNTHIA-TS, we have run an extensive set of experiments. The results show that the proposed method is indeed effective, provided that we use a proper Convolutional Neural Network (CNN) to perform the traffic sign recognition (classification) task as well as a proper GAN to transform the synthetic images. Here, a ResNet101-based classifier and domain adaptation based on CycleGAN performed extremely well for a ratio∼ 1/4 for new/known classes; even for more challenging ratios such as∼ 4/1, the results are also very positive.

Abstract: Nutrition and social interactions are both key aspects of the daily lives of humans. In this work, we propose a system to evaluate the influence of social interaction in the nutritional habits of a person from a first-person perspective. In order to detect the routine of an individual, we construct a nutritional behaviour pattern discovery model, which outputs routines over a number of days. Our method evaluates similarity of routines with respect to visited food-related scenes over the collected days, making use of Dynamic Time Warping, as well as considering social engagement and its correlation with food-related activities. The nutritional and social descriptors of the collected days are evaluated and encoded using an LSTM Autoencoder. Later, the obtained latent space is clustered to find similar days unaffected by outliers using the Isolation Forest method. Moreover, we introduce a new score metric to evaluate the performance of the proposed algorithm. We validate our method on 104 days and more than 100 k egocentric images gathered by 7 users. Several different visualizations are evaluated for the understanding of the findings. Our results demonstrate good performance and applicability of our proposed model for social-related nutritional behaviour understanding. At the end, relevant applications of the model are discussed by analysing the discovered routine of particular individuals.

Abstract: Today, we witness the appearance of many lifelogging cameras that are able to capture the life of a person wearing the camera and which produce a large number of images everyday. Automatically characterizing the experience and extracting patterns of behavior of individuals from this huge collection of unlabeled and unstructured egocentric data present major challenges and require novel and efficient algorithmic solutions. The main goal of this work is to propose a new method to automatically assess day similarity from the lifelogging images of a person. We propose a technique to measure the similarity between images based on the Swain’s distance and generalize it to detect the similarity between daily visual data. To this purpose, we apply the dynamic time warping (DTW) combined with the Swain’s distance for final day similarity estimation. For validation, we apply our technique on the Egocentric Dataset of University of Barcelona (EDUB) of 4912 daily images acquired by four persons with preliminary encouraging results.

Abstract: Hand sign language recognition from video is a challenging research area in computer vision, which performance is affected by hand occlusion, fast hand movement, illumination changes, or background complexity, just to mention a few. In recent years, deep learning approaches have achieved state-of-the-art results in the field, though previous challenges are not completely solved. In this work, we propose a novel deep learning-based pipeline architecture for efficient automatic hand sign language recognition using Single Shot Detector (SSD), 2D Convolutional Neural Network (2DCNN), 3D Convolutional Neural Network (3DCNN), and Long Short-Term Memory (LSTM) from RGB input videos. We use a CNN-based model which estimates the 3D hand keypoints from 2D input frames. After that, we connect these estimated keypoints to build the hand skeleton by using midpoint algorithm. In order to obtain a more discriminative representation of hands, we project 3D hand skeleton into three views surface images. We further employ the heatmap image of detected keypoints as input for refinement in a stacked fashion. We apply 3DCNNs on the stacked features of hand, including pixel level, multi-view hand skeleton, and heatmap features, to extract discriminant local spatio-temporal features from these stacked inputs. The outputs of the 3DCNNs are fused and fed to a LSTM to model long-term dynamics of hand sign gestures. Analyzing 2DCNN vs. 3DCNN using different number of stacked inputs into the network, we demonstrate that 3DCNN better capture spatio-temporal dynamics of hands. To the best of our knowledge, this is the first time that this multi-modal and multi-view set of hand skeleton features are applied for hand sign language recognition. Furthermore, we present a new large-scale hand sign language dataset, namely RKS-PERSIANSIGN, including 10′000 RGB videos of 100 Persian sign words. Evaluation results of the proposed model on three datasets, NYU, First-Person, and RKS-PERSIANSIGN, indicate that our model outperforms state-of-the-art models in hand sign language recognition, hand pose estimation, and hand action recognition.

Abstract: First impressions strongly influence social interactions, having a high impact in the personal and professional life. In this paper, we present a deep Classification-Regression Network (CR-Net) for analyzing the Big Five personality problem and further assisting on job interview recommendation in a first impressions setup. The setup is based on the ChaLearn First Impressions dataset, including multimodal data with video, audio, and text converted from the corresponding audio data, where each person is talking in front of a camera. In order to give a comprehensive prediction, we analyze the videos from both the entire scene (including the person’s motions and background) and the face of the person. Our CR-Net first performs personality trait classification and applies a regression later, which can obtain accurate predictions for both personality traits and interview recommendation. Furthermore, we present a new loss function called Bell Loss to address inaccurate predictions caused by the regression-to-the-mean problem. Extensive experiments on the First Impressions dataset show the effectiveness of our proposed network, outperforming the state-of-the-art.

Abstract: In handball videos recorded during the training, multiple players are present in the scene at the same time. Although they all might move and interact, not all players contribute to the currently relevant exercise nor practice the given handball techniques. The goal of this experiment is to automatically determine players on training footage that perform given handball techniques and are therefore considered active. It is a very challenging task for which a precise object detector is needed that can handle cluttered scenes with poor illumination, with many players present in different sizes and distances from the camera, partially occluded, moving fast. To determine which of the detected players are active, additional information is needed about the level of player activity. Since many handball actions are characterized by considerable changes in speed, position, and variations in the player's appearance, we propose using spatio-temporal interest points (STIPs) and optical flow (OF). Therefore, we propose an active player detection method combining the YOLO object detector and two activity measures based on STIPs and OF. The performance of the proposed method and activity measures are evaluated on a custom handball video dataset acquired during handball training lessons.

Abstract: Fine-grained image recognition is central to many multimedia tasks such as search, retrieval, and captioning. Unfortunately, these tasks are still challenging since the appearance of samples of the same class can be more different than those from different classes. This issue is mainly due to changes in deformation, pose, and the presence of clutter. In the literature, attention has been one of the most successful strategies to handle the aforementioned problems. Attention has been typically implemented in neural networks by selecting the most informative regions of the image that improve classification. In contrast, in this paper, attention is not applied at the image level but to the convolutional feature activations. In essence, with our approach, the neural model learns to attend to lower-level feature activations without requiring part annotations and uses those activations to update and rectify the output likelihood distribution. The proposed mechanism is modular, architecture-independent, and efficient in terms of both parameters and computation required. Experiments demonstrate that well-known networks such as wide residual networks and ResNeXt, when augmented with our approach, systematically improve their classification accuracy and become more robust to changes in deformation and pose and to the presence of clutter. As a result, our proposal reaches state-of-the-art classification accuracies in CIFAR-10, the Adience gender recognition task, Stanford Dogs, and UEC-Food100 while obtaining competitive performance in ImageNet, CIFAR-100, CUB200 Birds, and Stanford Cars. In addition, we analyze the different components of our model, showing that the proposed attention modules succeed in finding the most discriminative regions of the image. Finally, as a proof of concept, we demonstrate that with only local predictions, an augmented neural network can successfully classify an image before reaching any fully connected layer, thus reducing the computational amount up to 10%.