Showing papers by "Stefan Winkler published in 2020"
TL;DR: Zhang et al. as mentioned in this paper proposed an end-to-end trainable and deep Siamese-like network, which consists of two convolutional network branches, one for emotion and the other for apparent personality.
Abstract: Apparent personality and emotion analysis are both central to affective computing. Existing works solve them individually. In this paper we investigate if such high-level affect traits and their relationship can be jointly learned from face images in the wild. To this end, we introduce PersEmoN, an end-to-end trainable and deep Siamese-like network. It consists of two convolutional network branches, one for emotion and the other for apparent personality. Both networks share their bottom feature extraction module and are optimized within a multi-task learning framework. Emotion and personality networks are dedicated to their own annotated dataset. Furthermore, an adversarial-like loss function is employed to promote representation coherence among heterogeneous dataset sources. Based on this, we also explore the emotion-to-apparent-personality relationship. Extensive experiments demonstrate the effectiveness of PersEmoN.
27 citations
TL;DR: In this article, the efficacy of content-centric convolutional neural network (CNN) features for ad AR vis-a-vis handcrafted audio-visual descriptors was explored.
Abstract: Advertisements (ads) often contain strong emotions to capture audience attention and convey an effective message. Still, little work has focused on affect recognition (AR) from ads employing audiovisual or user cues. This work (1) compiles an affective video ad dataset which evokes coherent emotions across users; (2) explores the efficacy of content-centric convolutional neural network (CNN) features for ad AR vis-a-vis handcrafted audio-visual descriptors; (3) examines user-centric ad AR from Electroencephalogram (EEG) signals, and (4) demonstrates how better affect predictions facilitate effective computational advertising via a study involving 18 users. Experiments reveal that (a) CNN features outperform handcrafted audiovisual descriptors for content-centric AR; (b) EEG features encode ad-induced emotions better than contentbased features; (c) Multi-task learning achieves optimal ad AR among a slew of classifiers and (d) Pursuant to (b), EEG features enable optimized ad insertion onto streamed video compared to content-based or manual insertion, maximizing ad recall and viewing experience.
17 citations
03 Nov 2020
TL;DR: It is shown that when stochasticity is removed from the training procedure, GANs can overfit and exhibit almost no mode drop, providing evidence against prevailing intuitions that GAns do not memorize the training set, and that mode dropping is mainly due to properties of the GAN objective rather than how it is optimized during training.
Abstract: We examine two key questions in GAN training, namely overfitting and mode drop, from an empirical perspective. We show that when stochasticity is removed from the training procedure, GANs can overfit and exhibit almost no mode drop. Our results shed light on important characteristics of the GAN training procedure. They also provide evidence against prevailing intuitions that GANs do not memorize the training set, and that mode dropping is mainly due to properties of the GAN objective rather than how it is optimized during training.
17 citations
Posted Content•
TL;DR: Long-Short-Term-Memory networks have been used extensively for time series forecasting in recent years due to their ability of learning patterns over different periods of time, and this ability is applied to learning the pattern of Global Positioning System-based Precipitable Water Vapor measurements over a period of 4 hours.
Abstract: Long-Short-Term-Memory (LSTM) networks have been used extensively for time series forecasting in recent years due to their ability of learning patterns over different periods of time In this paper, this ability is applied to learning the pattern of Global Positioning System (GPS)-based Precipitable Water Vapor (PWV) measurements over a period of 4 hours The trained model was evaluated on more than 1500 hours of recorded data It achieves a root mean square error (RMSE) of 0098 mm for a forecasting interval of 5 minutes in the future, and outperforms the naive approach for a lead-time of up to 40 minutes
11 citations
05 Jul 2020
TL;DR: In this paper, a LSTM network was applied to learn the pattern of Global Positioning System (GPS)-based Precipitable Water Vapor (PWV) measurements over a period of 4 hours.
Abstract: Long-Short-Term-Memory (LSTM) networks have been used extensively for time series forecasting in recent years due to their ability of learning patterns over different periods of time In this paper, this ability is applied to learning the pattern of Global Positioning System (GPS)-based Precipitable Water Vapor (PWV) measurements over a period of 4 hours The trained model was evaluated on more than 1500 hours of recorded data It achieves a root mean square error (RMSE) of 0098mm for a forecasting interval of 5 minutes in the future, and outperforms the naive approach for a lead-time of up to 40 minutes
7 citations
TL;DR: In this article, the authors describe the design of new readout chambers and front-end electronics, which are driven by the goals of the ALICE TPC at the CERN LHC.
Abstract: The upgrade of the ALICE TPC will allow the experiment to cope with the high interaction rates foreseen for the forthcoming Run 3 and Run 4 at the CERN LHC. In this article, we describe the design of new readout chambers and front-end electronics, which are driven by the goals of the experiment. Gas Electron Multiplier (GEM) detectors arranged in stacks containing four GEMs each, and continuous readout electronics based on the SAMPA chip, an ALICE development, are replacing the previous elements. The construction of these new elements, together with their associated quality control procedures, is explained in detail. Finally, the readout chamber and front-end electronics cards replacement, together with the commissioning of the detector prior to installation in the experimental cavern, are presented. After a nine-year period of R&D, construction, and assembly, the upgrade of the TPC was completed in 2020.
4 citations
02 Nov 2020
TL;DR: A frontalization technique for 2D facial landmarks, designed to aid in the analysis of facial expressions, employs a new normalization strategy aiming to minimize identity variations, by displacing groups of facial landmarks to standardized locations.
Abstract: We propose a frontalization technique for 2D facial landmarks, designed to aid in the analysis of facial expressions. It employs a new normalization strategy aiming to minimize identity variations, by displacing groups of facial landmarks to standardized locations. The technique operates directly on 2D landmark coordinates, does not require additional feature extraction and as such is computationally light. It achieves considerable improvement over a reference approach, justifying its use as an efficient preprocessing step for facial expression analysis based on geometric features.
3 citations
Posted Content•
TL;DR: In this paper, the authors examine two key questions in GAN training, namely overfitting and mode drop, from an empirical perspective and show that when stochasticity is removed from the training procedure, GANs exhibit almost no mode drop.
Abstract: We examine two key questions in GAN training, namely overfitting and mode drop, from an empirical perspective We show that when stochasticity is removed from the training procedure, GANs can overfit and exhibit almost no mode drop Our results shed light on important characteristics of the GAN training procedure They also provide evidence against prevailing intuitions that GANs do not memorize the training set, and that mode dropping is mainly due to properties of the GAN objective rather than how it is optimized during training
1 citations
16 Nov 2020
TL;DR: In this article, a dataset of aerial images of helipads, together with a method to identify and locate such helicopter bases from the air, is presented, including a classification by visual helipad shape and features, which makes available to the research community.
Abstract: We present HelipadCat, a dataset of aerial images of helipads, together with a method to identify and locate such helipads from the air. Based on the FAA’s database of US airports, we create the first dataset of helipads, including a classification by visual helipad shape and features, which we make available to the research community. The dataset includes nearly 6,000 images with 12 different categories.We then train several Mask-RCNN models based on ResNet101 using our dataset. Image augmentation is applied according to learned augmentation policies. We characterize the performance of the models on HelipadCat and pick the best-performing configuration. We further evaluate that model on the metropolitan area of Manila and show that it is able to detect helipads successfully, with their exact geographical coordinates, in another country. To reduce false positives, the bounding boxes are filtered by confidence score, size, and the presence of shadows. Dataset and code are available for download.