Contrastive learning constitutes an emerging branch of self-supervised learning that leverages large amounts of unlabeled data, by learning a latent space, where pairs of different views of the same sample are associated. In this paper, we propose musical source association as a pair generation strategy in the context of contrastive music representation learning. To this end, we modify COLA, a widely used contrastive learning audio framework, to learn to associate a song excerpt with a stochastically selected and automatically extracted vocal or instrumental source. We further introduce a novel modification to the contrastive loss to incorporate information about the existence or absence of specific sources. Our experimental evaluation in three different downstream tasks (music auto-tagging, instrument classification and music genre classification) using the publicly available Magna-Tag-A-Tune (MTAT) as a source dataset yields competitive results to existing literature methods, as well as faster network convergence. The results also show that this pre-training method can be steered towards specific features, according to the selected musical source, while also being dependent on the quality of the separated sources.
Traditional monocular Visual Simultaneous Localization and Mapping (vSLAM) systems can be divided into three categories: those that use features, those that rely on the image itself, and hybrid models. In the case of feature-based methods, new research has evolved to incorporate more information from their environment using geometric primitives beyond points, such as lines and planes. This is because in many environments, which are man-made environments, characterized as Manhattan world, geometric primitives such as lines and planes occupy most of the space in the environment. The exploitation of these schemes can lead to the introduction of algorithms capable of optimizing the trajectory of a Visual SLAM system and also helping to construct an exuberant map. Thus, we present a real-time monocular Visual SLAM system that incorporates real-time methods for line and VP extraction, as well as two strategies that exploit vanishing points to estimate the robot's translation and improve its rotation.Particularly, we build on ORB-SLAM2, which is considered the current state-of-the-art solution in terms of both accuracy and efficiency, and extend its formulation to handle lines and VPs to create two strategies the first optimize the rotation and the second refine the translation part from the known rotation. First, we extract VPs using a real-time method and use them for a global rotation optimization strategy. Second, we present a translation estimation method that takes advantage of last-stage rotation optimization to model a linear system. Finally, we evaluate our system on the TUM RGB-D benchmark and demonstrate that the proposed system achieves state-of-the-art results and runs in real time, and its performance remains close to the original ORB-SLAM2 system
In recent years, implicit surface representations through neural networks that encode the signed distance have gained popularity and have achieved state-of-the-art results in various tasks (e.g. shape representation, shape reconstruction, and learning shape priors). However, in contrast to conventional shape representations such as polygon meshes, the implicit representations cannot be easily edited and existing works that attempt to address this problem are extremely limited. In this work, we propose the first method for efficient interactive editing of signed distance functions expressed through neural networks, allowing free-form editing. Inspired by 3D sculpting software for meshes, we use a brush-based framework that is intuitive and can in the future be used by sculptors and digital artists. In order to localize the desired surface deformations, we regulate the network by using a copy of it to sample the previously expressed surface. We introduce a novel framework for simulating sculpting-style surface edits, in conjunction with interactive surface sampling and efficient adaptation of network weights. We qualitatively and quantitatively evaluate our method in various different 3D objects and under many different edits. The reported results clearly show that our method yields high accuracy, in terms of achieving the desired edits, while at the same time preserving the geometry outside the interaction areas.
In this paper, we introduce a neural rendering pipeline for transferring the facial expressions, head pose and body movements of one person in a source video to another in a target video. We apply our method to the challenging case of Sign Language videos: given a source video of a sign language user, we can faithfully transfer the performed manual (e.g. handshape, palm orientation, movement, location) and non-manual (e.g. eye gaze, facial expressions, head movements) signs to a target video in a photo-realistic manner. To effectively capture the aforementioned cues, which are crucial for sign language communication, we build upon an effective combination of the most robust and reliable deep learning methods for body, hand and face tracking that have been introduced lately. Using a 3D-aware representation, the estimated motions of the body parts are combined and retargeted to the target signer. They are then given as conditional input to our Video Rendering Network, which generates temporally consistent and photo-realistic videos. We conduct detailed qualitative and quantitative evaluations and comparisons, which demonstrate the effectiveness of our approach and its advantages over existing approaches. Our method yields promising results of unprecedented realism and can be used for Sign Language Anonymization. In addition, it can be readily applicable to reenactment of other types of full body activities (dancing, acting performance, exercising, etc.), as well as to the synthesis module of Sign Language Production systems.
The recent state of the art on monocular 3D face reconstruction from image data has made some impressive advancements, thanks to the advent of Deep Learning. However, it has mostly focused on input coming from a single RGB image, overlooking the following important factors: a) Nowadays, the vast majority of facial image data of interest do not originate from single images but rather from videos, which contain rich dynamic information. b) Furthermore, these videos typically capture individuals in some form of verbal communication (public talks, teleconferences, audiovisual human-computer interactions, interviews, monologues/dialogues in movies, etc). When existing 3D face reconstruction methods are applied in such videos, the artifacts in the reconstruction of the shape and motion of the mouth area are often severe, since they do not match well with the speech audio. To overcome the aforementioned limitations, we present the first method for visual speech-aware perceptual reconstruction of 3D mouth expressions. We do this by proposing a "lipread" loss, which guides the fitting process so that the elicited perception from the 3D reconstructed talking head resembles that of the original video footage. We demonstrate that, interestingly, the lipread loss is better suited for 3D reconstruction of mouth movements compared to traditional landmark losses, and even direct 3D supervision. Furthermore, the devised method does not rely on any text transcriptions or corresponding audio, rendering it ideal for training in unlabeled datasets. We verify the efficiency of our method through exhaustive objective evaluations on three large-scale datasets, as well as subjective evaluation with two web-based user studies.
The study of Music Cognition and neural responses to music has been invaluable in understanding human emotions. Brain signals, though, manifest a highly complex structure that makes processing and retrieving meaningful features challenging, particularly of abstract constructs like affect. Moreover, the performance of learning models is undermined by the limited amount of available neuronal data and their severe inter-subject variability. In this paper we extract efficient, personalized affective representations from EEG signals during music listening. To this end, we employ music signals as a supervisory modality to EEG, aiming to project their semantic correspondence onto a common representation space. We utilize a bi-modal framework by combining an LSTM-based attention model to process EEG and a pre-trained model for music tagging, along with a reverse domain discriminator to align the distributions of the two modalities, further constraining the learning process with emotion tags. The resulting framework can be utilized for emotion recognition both directly, by performing supervised predictions from either modality, and indirectly, by providing relevant music samples to EEG input queries. The experimental findings show the potential of enhancing neuronal data through stimulus information for recognition purposes and yield insights into the distribution and temporal variance of music-induced affective features.
In this paper, we introduce a novel deep learning method for photo-realistic manipulation of the emotional state of actors in "in-the-wild" videos. The proposed method is based on a parametric 3D face representation of the actor in the input scene that offers a reliable disentanglement of the facial identity from the head pose and facial expressions. It then uses a novel deep domain translation framework that alters the facial expressions in a consistent and plausible manner, taking into account their dynamics. Finally, the altered facial expressions are used to photo-realistically manipulate the facial region in the input scene based on an especially-designed neural face renderer. To the best of our knowledge, our method is the first to be capable of controlling the actor's facial expressions by even using as a sole input the semantic labels of the manipulated emotions, while at the same time preserving the speech-related lip movements. We conduct extensive qualitative and quantitative evaluations and comparisons, which demonstrate the effectiveness of our approach and the especially promising results that we obtain. Our method opens a plethora of new possibilities for useful applications of neural rendering technologies, ranging from movie post-production and video games to photo-realistic affective avatars.
In this work we tackle the task of video-based audio-visual emotion recognition, within the premises of the 2nd Workshop and Competition on Affective Behavior Analysis in-the-wild (ABAW). Poor illumination conditions, head/body orientation and low image resolution constitute factors that can potentially hinder performance in case of methodologies that solely rely on the extraction and analysis of facial features. In order to alleviate this problem, we leverage bodily as well as contextual features, as part of a broader emotion recognition framework. We choose to use a standard CNN-RNN cascade as the backbone of our proposed model for sequence-to-sequence (seq2seq) learning. Apart from learning through the RGB input modality, we construct an aural stream which operates on sequences of extracted mel-spectrograms. Our extensive experiments on the challenging and newly assembled Affect-in-the-wild-2 (Aff-Wild2) dataset verify the superiority of our methods over existing approaches, while by properly incorporating all of the aforementioned modules in a network ensemble, we manage to surpass the previous best published recognition scores, in the official validation set. All the code was implemented using PyTorch\footnote{\url{https://pytorch.org/}} and is publicly available\footnote{\url{https://github.com/PanosAntoniadis/NTUA-ABAW2021}}.
Nowadays, the interaction between humans and robots is constantly expanding, requiring more and more human motion recognition applications to operate in real time. However, most works on temporal action detection and recognition perform these tasks in offline manner, i.e. temporally segmented videos are classified as a whole. In this paper, based on the recently proposed framework of Temporal Recurrent Networks, we explore how temporal context and human movement dynamics can be effectively employed for online action detection. Our approach uses various state-of-the-art architectures and appropriately combines the extracted features in order to improve action detection. We evaluate our method on a challenging but widely used dataset for temporal action localization, THUMOS'14. Our experiments show significant improvement over the baseline method, achieving state-of-the art results on THUMOS'14.