Adaptive Fusion Techniques for Multimodal Data

Proceedings of the 22nd Conference on Computational Natural Language Learning

Natural human communication is nuanced and inherently multi-modal. Humans possess specialised sensoria for processing vocal, visual, and linguistic, and para-linguistic information, but form an intricately fused percept of the multi-modal data stream to provide a holistic representation. Analysis of emotional content in face-to-face communication is a cognitive task to which humans are particularly attuned, given its sociological importance, and poses a difficult challenge for machine emulation due to the subtlety and expressive variability of cross-modal cues. Inspired by the empirical success of recent so-called End-To-End Memory Networks (Sukhbaatar et al., 2015), we propose an approach based on recursive multi-attention with a shared external memory updated over multiple gated iterations of analysis. We evaluate our model across several large multimodal datasets and show that global contextualised memory with gated memory update can effectively achieve emotion recognition.

ArXiv, 2019

This paper presents a novel deep neural network (DNN) for multimodal fusion of audio, video and text modalities for emotion recognition. The proposed DNN architecture has independent and shared layers which aim to learn the representation for each modality, as well as the best combined representation to achieve the best prediction. Experimental results on the AVEC Sentiment Analysis in the Wild dataset indicate that the proposed DNN can achieve a higher level of Concordance Correlation Coefficient (CCC) than other state-of-the-art systems that perform early fusion of modalities at feature-level (i.e., concatenation) and late fusion at score-level (i.e., weighted average) fusion. The proposed DNN has achieved CCCs of 0.606, 0.534, and 0.170 on the development partition of the dataset for predicting arousal, valence and liking, respectively.

Proceedings of Grand Challenge and Workshop on Human Multimodal Language (Challenge-HML), 2018

We present our system description of input-level multimodal fusion of audio, video, and text for recognition of emotions and their intensities for the 2018 First Grand Challenge on Computational Modeling of Human Multimodal Language. Our proposed approach is based on input-level feature fusion with sequence learning from Bidirectional Long-Short Term Memory (BLSTM) deep neural networks (DNNs). We show that our fusion approach outperforms unimodal predictors. Our system performs 6-way simultaneous classification and regression, allowing for overlapping emotion labels in a video segment. This leads to an overall binary accuracy of 90%, overall 4-class accuracy of 89.2% and an overall mean-absolute-error (MAE) of 0.12. Our work shows that an early fusion technique can effectively predict the presence of multi-label emotions as well as their coarsegrained intensities. The presented multimodal approach creates a simple and robust baseline on this new Grand Challenge dataset. Furthermore, we provide a detailed analysis of emotion intensity distributions as output from our DNN, as well as a related discussion concerning the inherent difficulty of this task.

In this paper, we describe our work in the third Emotion Recognition in the Wild (EmotiW 2015) Challenge. For each video clip, we extract MSDF, LBP-TOP, HOG, LPQ-TOP and acoustic features to recognize the emotions of film characters. For the static facial expression recognition based on video frame, we extract MSDF, DCNN and RCNN features. We train linear SVM classifiers for these kinds of features on the AFEW and SFEW dataset, and we propose a novel fusion network to combine all the extracted features at decision level. The final achievement we gained is 51.02% on the AFEW testing set and 51.08% on the SFEW testing set, which are much better than the baseline recognition rate of 39.33% and 39.13%.

2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)

Emotion Recognition in Conversations (ERC) is crucial in developing sympathetic human-machine interaction. In conversational videos, emotion can be present in multiple modalities, i.e., audio, video, and transcript. However, due to the inherent characteristics of these modalities, multi-modal ERC has always been considered a challenging undertaking. Existing ERC research focuses mainly on using text information in a discussion, ignoring the other two modalities. We anticipate that emotion recognition accuracy can be improved by employing a multi-modal approach. Thus, in this study, we propose a Multi-modal Fusion Network (M2FNet) that extracts emotion-relevant features from visual, audio, and text modality. It employs a multi-head attention-based fusion mechanism to combine emotion-rich latent representations of the input data. We introduce a new feature extractor to extract latent features from the audio and visual modality. The proposed feature extractor is trained with a novel adaptive marginbased triplet loss function to learn emotion-relevant features from the audio and visual data. In the domain of ERC, the existing methods perform well on one benchmark dataset but not on others. Our results show that the proposed M2FNet architecture outperforms all other methods in terms of weighted average F1 score on well-known MELD and IEMOCAP datasets and sets a new state-of-theart performance in ERC.

2020 IEEE International Conference on Data Mining (ICDM), 2020

Multimodal sentiment analysis utilizes multiple heterogeneous modalities for sentiment classification. The recent multimodal fusion schemes customize LSTMs to discover intramodal dynamics and design sophisticated attention mechanisms to discover the intermodal dynamics from multimodal sequences. Although powerful, these schemes completely rely on attention mechanisms which is problematic due to two major drawbacks 1) deceptive attention masks, and 2) training dynamics. Nevertheless, strenuous efforts are required to optimize hyperparameters of these consolidate architectures, in particular their custom-designed LSTMs constrained by attention schemes. In this research, we first propose a common network to discover both intra-modal and intermodal dynamics by utilizing basic LSTMs and tensor based convolution networks. We then propose unique networks to encapsulate temporal-granularity among the modalities which is essential while extracting information within asynchronous sequences. We then integrate these two kinds of information via a fusion layer and call our novel multimodal fusion scheme as Deep-HOSeq (Deep network with higher order Common and Unique Sequence information). The proposed Deep-HOSeq efficiently discovers all-important information from multimodal sequences and the effectiveness of utilizing both types of information is empirically demonstrated on CMU-MOSEI and CMU-MOSI benchmark datasets. The source code of our proposed Deep-HOSeq is and available at https://github.com/sverma88/Deep-HOSeq-ICDM-2020. Index Terms-multimodal data fusion, sentiment analysis, tensor analysis, convolution neural networks.

IEEE Transactions on Signal Processing, 2021

The commonly used latent space embedding techniques, such as Principal Component Analysis, Factor Analysis, and manifold learning techniques, are typically used for learning effective representations of homogeneous data. However, they do not readily extend to heterogeneous data that are a combination of numerical and categorical variables, e.g., arising from linked GPS and text data. In this paper, we are interested in learning probabilistic generative models from high-dimensional heterogeneous data in an unsupervised fashion. The learned generative model provides latent unified representations that capture the factors common to the multiple dimensions of the data, and thus enable fusing multimodal data for various machine learning tasks. Following a Bayesian approach, we propose a general framework that combines disparate data types through the natural parameterization of the exponential family of distributions. To scale the model inference to millions of instances with thousands of features, we use the Laplace-Bernstein approximation for posterior computations involving nonlinear link functions. The proposed algorithm is presented in detail for the commonly encountered heterogeneous datasets with real-valued (Gaussian) and categorical (multinomial) features. Experiments on two high-dimensional and heterogeneous datasets (NYC Taxi and MovieLens-10M) demonstrate the scalability and competitive performance of the proposed algorithm on different machine learning tasks such as anomaly detection, data imputation, and recommender systems.

Proceedings of the 2nd on Multimodal Sentiment Analysis Challenge

Human stress detection is of great importance for monitoring mental health. The Multimodal Sentiment Analysis Challenge (MuSe) 2021 focuses on emotion, physiological-emotion, and stress recognition as well as sentiment classification by exploiting several modalities. In this paper, we present our solution for the Muse-Stress sub-challenge. The target of this sub-challenge is continuous prediction of arousal and valence for people under stressful conditions where text transcripts, audio and video recordings are provided. To this end, we utilize bidirectional Long Short-Term Memory (LSTM) and Gated Recurrent Unit networks (GRU) to explore high-level and lowlevel features from different modalities. We employ Concordance Correlation Coefficient (CCC) as a loss function and evaluation metric for our model. To improve the unimodal predictions, we add difficulty indicators of the data obtained by using Auto-Encoders. Finally, we perform late fusion on our unimodal predictions in addition to the difficulty indicators to obtain our final predictions. With this approach, we achieve CCC of 0.4278 and 0.5951 for arousal and valence respectively on the test set, our submission to MuSe 2021 ranks in the top three for arousal, fourth for valence, and in top three for combined results.

International Journal of Emerging Trends in Engineering Research, 2024

Emotion recognition is a pivotal area of research with applications spanning education, healthcare, and intelligent customer service. Multimodal emotion recognition (MER) has emerged as a superior approach by integrating multiple modalities such as speech, text, and facial expressions, offering enhanced accuracy and robustness over unimodal methods. This paper reviews the evolution and current state of MER, highlighting its significance, challenges, and methodologies. We delve into various datasets, including IEMOCAP and MELD, providing a comparative analysis of their strengths and limitations. The literature review covers recent advancements in deep learning techniques, focusing on fusion strategies like early, late, and hybrid fusion. Identified gaps include issues related to data redundancy, feature extraction complexity, and real-time detection. Our proposed methodology leverages deep learning for feature extraction and a hybrid fusion approach to improve emotion detection accuracy. This research aims to guide future studies in addressing current limitations and advancing the field of MER. The main of this paper review recent methodologies in multimodal emotion recognition, analyze different data fusion techniques,identify challenges and research gaps.

arXiv (Cornell University), 2022

Automatically recognising apparent emotions from face and voice is hard, in part because of various sources of uncertainty, including in the input data and the labels used in a machine learning framework. This paper introduces an uncertainty-aware multimodal fusion approach that quantifies modality-wise aleatoric or data uncertainty towards emotion prediction. We propose a novel fusion framework, in which latent distributions over unimodal temporal context are learned by constraining their variance. These variance constraints, Calibration and Ordinal Ranking, are designed such that the variance estimated for a modality can represent how informative the temporal context of that modality is w.r.t. emotion recognition. When well-calibrated, modality-wise uncertainty scores indicate how much their corresponding predictions are likely to differ from the ground truth labels. Well-ranked uncertainty scores allow the ordinal ranking of different frames across different modalities. To jointly impose both these constraints, we propose a softmax distributional matching loss. Our evaluation on AVEC 2019 CES, CMU-MOSEI, and IEMOCAP datasets shows that the proposed multimodal fusion method not only improves the generalisation performance of emotion recognition models and their predictive uncertainty estimates, but also makes the models robust to novel noise patterns encountered at test time.