Search

We present the Capsule Vision 2024 Challenge: Multi-Class Abnormality Classification for Video Capsule Endoscopy. It was virtually organized by the Research Center for Medical Image Analysis and Artificial Intelligence (MIAAI), Department of Medicine, Danube Private University, Krems, Austria in collaboration with the 9th International Conference on Computer Vision & Image Processing (CVIP 2024) being organized by the Indian Institute of Information Technology, Design and Manufacturing (IIITDM) Kancheepuram, Chennai, India. This document provides an overview of the challenge, including the registration process, rules, submission format, description of the datasets used, qualified team rankings, all team descriptions, and the benchmarking results reported by the organizers., Comment: 11 pages

Deep learning models, particularly Convolutional Neural Networks (CNNs), have demonstrated exceptional performance in diagnosing skin diseases, often outperforming dermatologists. However, they have also unveiled biases linked to specific demographic traits, notably concerning diverse skin tones or gender, prompting concerns regarding fairness and limiting their widespread deployment. Researchers are actively working to ensure fairness in AI-based solutions, but existing methods incur an accuracy loss when striving for fairness. To solve this issue, we propose a `two-biased teachers' (i.e., biased on different sensitive attributes) based approach to transfer fair knowledge into the student network. Our approach mitigates biases present in the student network without harming its predictive accuracy. In fact, in most cases, our approach improves the accuracy of the baseline model. To achieve this goal, we developed a weighted loss function comprising biasing and debiasing loss terms. We surpassed available state-of-the-art approaches to attain fairness and also improved the accuracy at the same time. The proposed approach has been evaluated and validated on two dermatology datasets using standard accuracy and fairness evaluation measures. We will make source code publicly available to foster reproducibility and future research.

This paper introduces a new stochastic optimization method based on the regularized Fisher information matrix (FIM), named SOFIM, which can efficiently utilize the FIM to approximate the Hessian matrix for finding Newton's gradient update in large-scale stochastic optimization of machine learning models. It can be viewed as a variant of natural gradient descent, where the challenge of storing and calculating the full FIM is addressed through making use of the regularized FIM and directly finding the gradient update direction via Sherman-Morrison matrix inversion. Additionally, like the popular Adam method, SOFIM uses the first moment of the gradient to address the issue of non-stationary objectives across mini-batches due to heterogeneous data. The utilization of the regularized FIM and Sherman-Morrison matrix inversion leads to the improved convergence rate with the same space and time complexities as stochastic gradient descent (SGD) with momentum. The extensive experiments on training deep learning models using several benchmark image classification datasets demonstrate that the proposed SOFIM outperforms SGD with momentum and several state-of-the-art Newton optimization methods in term of the convergence speed for achieving the pre-specified objectives of training and test losses as well as test accuracy.

The ability to generate sentiment-controlled feedback in response to multimodal inputs comprising text and images addresses a critical gap in human-computer interaction. This capability allows systems to provide empathetic, accurate, and engaging responses, with useful applications in education, healthcare, marketing, and customer service. To this end, we have constructed a large-scale Controllable Multimodal Feedback Synthesis (CMFeed) dataset and propose a controllable feedback synthesis system. The system features an encoder, decoder, and controllability block for textual and visual inputs. It extracts features using a transformer and Faster R-CNN networks, combining them to generate feedback. The CMFeed dataset includes images, texts, reactions to the posts, human comments with relevance scores, and reactions to these comments. These reactions train the model to produce feedback with specified sentiments, achieving a sentiment classification accuracy of 77.23\%, which is 18.82\% higher than the accuracy without controllability. The system also incorporates a similarity module for assessing feedback relevance through rank-based metrics and an interpretability technique to analyze the contributions of textual and visual features during feedback generation. Access to the CMFeed dataset and the system's code is available at https://github.com/MIntelligence-Group/CMFeed.

In the rapidly evolving landscape of medical imaging diagnostics, achieving high accuracy while preserving computational efficiency remains a formidable challenge. This work presents \texttt{DeepMediX}, a groundbreaking, resource-efficient model that significantly addresses this challenge. Built on top of the MobileNetV2 architecture, DeepMediX excels in classifying brain MRI scans and skin cancer images, with superior performance demonstrated on both binary and multiclass skin cancer datasets. It provides a solution to labor-intensive manual processes, the need for large datasets, and complexities related to image properties. DeepMediX's design also includes the concept of Federated Learning, enabling a collaborative learning approach without compromising data privacy. This approach allows diverse healthcare institutions to benefit from shared learning experiences without the necessity of direct data access, enhancing the model's predictive power while preserving the privacy and integrity of sensitive patient data. Its low computational footprint makes DeepMediX suitable for deployment on handheld devices, offering potential for real-time diagnostic support. Through rigorous testing on standard datasets, including the ISIC2018 for dermatological research, DeepMediX demonstrates exceptional diagnostic capabilities, matching the performance of existing models on almost all tasks and even outperforming them in some cases. The findings of this study underline significant implications for the development and deployment of AI-based tools in medical imaging and their integration into point-of-care settings. The source code and models generated would be released at https://github.com/kishorebabun/DeepMediX., Comment: 23 pages, 3 figures, 4 tables, 1 algorithm

The proliferation of deep learning applications in healthcare calls for data aggregation across various institutions, a practice often associated with significant privacy concerns. This concern intensifies in medical image analysis, where privacy-preserving mechanisms are paramount due to the data being sensitive in nature. Federated learning, which enables cooperative model training without direct data exchange, presents a promising solution. Nevertheless, the inherent vulnerabilities of federated learning necessitate further privacy safeguards. This study addresses this need by integrating differential privacy, a leading privacy-preserving technique, into a federated learning framework for medical image classification. We introduce a novel differentially private federated learning model and meticulously examine its impacts on privacy preservation and model performance. Our research confirms the existence of a trade-off between model accuracy and privacy settings. However, we demonstrate that strategic calibration of the privacy budget in differential privacy can uphold robust image classification performance while providing substantial privacy protection., Comment: 18 pages, 3 figures, 1 table, 1 algorithm

In recent years, deep learning models have revolutionized medical image interpretation, offering substantial improvements in diagnostic accuracy. However, these models often struggle with challenging images where critical features are partially or fully occluded, which is a common scenario in clinical practice. In this paper, we propose a novel curriculum learning-based approach to train deep learning models to handle occluded medical images effectively. Our method progressively introduces occlusion, starting from clear, unobstructed images and gradually moving to images with increasing occlusion levels. This ordered learning process, akin to human learning, allows the model to first grasp simple, discernable patterns and subsequently build upon this knowledge to understand more complicated, occluded scenarios. Furthermore, we present three novel occlusion synthesis methods, namely Wasserstein Curriculum Learning (WCL), Information Adaptive Learning (IAL), and Geodesic Curriculum Learning (GCL). Our extensive experiments on diverse medical image datasets demonstrate substantial improvements in model robustness and diagnostic accuracy over conventional training methodologies., Comment: 25 pages, 3 figures, 1 table (supplementary section added)

In healthcare, accurately classifying medical images is vital, but conventional methods often hinge on medical data with a consistent grid structure, which may restrict their overall performance. Recent medical research has been focused on tweaking the architectures to attain better performance without giving due consideration to the representation of data. In this paper, we present a novel approach for transforming grid based data into its higher dimensional representations, leveraging unstructured point cloud data structures. We first generate a sparse point cloud from an image by integrating pixel color information as spatial coordinates. Next, we construct a hypersurface composed of points based on the image dimensions, with each smooth section within this hypersurface symbolizing a specific pixel location. Polygonal face construction is achieved using an adjacency tensor. Finally, a dense point cloud is generated by densely sampling the constructed hypersurface, with a focus on regions of higher detail. The effectiveness of our approach is demonstrated on a publicly accessible brain tumor dataset, achieving significant improvements over existing classification techniques. This methodology allows the extraction of intricate details from the original image, opening up new possibilities for advanced image analysis and processing tasks.

This paper proposes a multimodal emotion recognition system based on hybrid fusion that classifies the emotions depicted by speech utterances and corresponding images into discrete classes. A new interpretability technique has been developed to identify the important speech & image features leading to the prediction of particular emotion classes. The proposed system's architecture has been determined through intensive ablation studies. It fuses the speech & image features and then combines speech, image, and intermediate fusion outputs. The proposed interpretability technique incorporates the divide & conquer approach to compute shapely values denoting each speech & image feature's importance. We have also constructed a large-scale dataset (IIT-R SIER dataset), consisting of speech utterances, corresponding images, and class labels, i.e., 'anger,' 'happy,' 'hate,' and 'sad.' The proposed system has achieved 83.29% accuracy for emotion recognition. The enhanced performance of the proposed system advocates the importance of utilizing complementary information from multiple modalities for emotion recognition., Comment: arXiv admin note: text overlap with arXiv:2208.11450

This paper proposes a multimodal emotion recognition system, VIsual Spoken Textual Additive Net (VISTANet), to classify emotions reflected by input containing image, speech, and text into discrete classes. A new interpretability technique, K-Average Additive exPlanation (KAAP), has been developed that identifies important visual, spoken, and textual features leading to predicting a particular emotion class. The VISTANet fuses information from image, speech, and text modalities using a hybrid of early and late fusion. It automatically adjusts the weights of their intermediate outputs while computing the weighted average. The KAAP technique computes the contribution of each modality and corresponding features toward predicting a particular emotion class. To mitigate the insufficiency of multimodal emotion datasets labeled with discrete emotion classes, we have constructed a large-scale IIT-R MMEmoRec dataset consisting of images, corresponding speech and text, and emotion labels ('angry,' 'happy,' 'hate,' and 'sad'). The VISTANet has resulted in 95.99% emotion recognition accuracy on the IIT-R MMEmoRec dataset using visual, audio, and textual modalities, outperforming when using any one or two modalities. The IIT-R MMEmoRec dataset can be accessed at https://github.com/MIntelligence-Group/MMEmoRec.

In this paper, we have defined a novel task of affective feedback synthesis that deals with generating feedback for input text & corresponding image in a similar way as humans respond towards the multimodal data. A feedback synthesis system has been proposed and trained using ground-truth human comments along with image-text input. We have also constructed a large-scale dataset consisting of image, text, Twitter user comments, and the number of likes for the comments by crawling the news articles through Twitter feeds. The proposed system extracts textual features using a transformer-based textual encoder while the visual features have been extracted using a Faster region-based convolutional neural networks model. The textual and visual features have been concatenated to construct the multimodal features using which the decoder synthesizes the feedback. We have compared the results of the proposed system with the baseline models using quantitative and qualitative measures. The generated feedbacks have been analyzed using automatic and human evaluation. They have been found to be semantically similar to the ground-truth comments and relevant to the given text-image input., Comment: Submitted to ACM Transactions on Multimedia Computing, Communications, and Applications

Location-Based Recommendation Services (LBRS) has seen an unprecedented rise in its usage in recent years. LBRS facilitates a user by recommending services based on his location and past preferences. However, leveraging such services comes at a cost of compromising one's sensitive information like their shopping preferences, lodging places, food habits, recently visited places, etc. to the third-party servers. Losing such information could be crucial and threatens one's privacy. Nowadays, the privacy-aware society seeks solutions that can provide such services, with minimized risks. Recently, a few privacy-preserving recommendation services have been proposed that exploit the fully homomorphic encryption (FHE) properties to address the issue. Though, it reduced privacy risks but suffered from heavy computational overheads that ruled out their commercial applications. Here, we propose SHELBRS, a lightweight LBRS that is based on switchable homomorphic encryption (SHE), which will benefit the users as well as the service providers. A SHE exploits both the additive as well as the multiplicative homomorphic properties but with comparatively much lesser processing time as it's FHE counterpart. We evaluate the performance of our proposed scheme with the other state-of-the-art approaches without compromising security., Comment: 9 pages, 6 figures, 3 tables

This paper has proposed a novel approach to classify the subjects' smoking behavior by extracting relevant regions from a given image using deep learning. After the classification, we have proposed a conditional detection module based on Yolo-v3, which improves model's performance and reduces its complexity. As per the best of our knowledge, we are the first to work on this dataset. This dataset contains a total of 2,400 images that include smokers and non-smokers equally in various environmental settings. We have evaluated the proposed approach's performance using quantitative and qualitative measures, which confirms its effectiveness in challenging situations. The proposed approach has achieved a classification accuracy of 96.74% on this dataset., Comment: 5 pages, 16 figures. To appear in the proceedings of the 28th IEEE International Conference on Image Processing (IEEE - ICIP), September 19-22, 2021, Anchorage, Alaska, USA

This paper proposes a feature-based domain adaptation technique for identifying emotions in generic images, encompassing both facial and non-facial objects, as well as non-human components. This approach addresses the challenge of the limited availability of pre-trained models and well-annotated datasets for Image Emotion Recognition (IER). Initially, a deep-learning-based Facial Expression Recognition (FER) system is developed, classifying facial images into discrete emotion classes. Maintaining the same network architecture, this FER system is then adapted to recognize emotions in generic images through the application of discrepancy loss, enabling the model to effectively learn IER features while classifying emotions into categories such as 'happy,' 'sad,' 'hate,' and 'anger.' Additionally, a novel interpretability method, Divide and Conquer based Shap (DnCShap), is introduced to elucidate the visual features most relevant for emotion recognition. The proposed IER system demonstrated emotion classification accuracies of 61.86% for the IAPSa dataset, 62.47 for the ArtPhoto dataset, 70.78% for the FI dataset, and 59.72% for the EMOTIC dataset. The system effectively identifies the important visual features that lead to specific emotion classifications and also provides detailed embedding plots explaining the predictions, enhancing the understanding and trust in AI-driven emotion recognition systems.

In this paper, an end-to-end neural embedding system based on triplet loss and residual learning has been proposed for speech emotion recognition. The proposed system learns the embeddings from the emotional information of the speech utterances. The learned embeddings are used to recognize the emotions portrayed by given speech samples of various lengths. The proposed system implements Residual Neural Network architecture. It is trained using softmax pre-training and triplet loss function. The weights between the fully connected and embedding layers of the trained network are used to calculate the embedding values. The embedding representations of various emotions are mapped onto a hyperplane, and the angles among them are computed using the cosine similarity. These angles are utilized to classify a new speech sample into its appropriate emotion class. The proposed system has demonstrated 91.67% and 64.44% accuracy while recognizing emotions for RAVDESS and IEMOCAP dataset, respectively., Comment: Accepted in ICPR 2020

Evaluation of students' performance for the completion of courses has been a major problem for both students and faculties during the work-from-home period in this COVID pandemic situation. To this end, this paper presents an in-depth analysis of deep learning and machine learning approaches for the formulation of an automated students' performance estimation system that works on partially available students' academic records. Our main contributions are (a) a large dataset with fifteen courses (shared publicly for academic research) (b) statistical analysis and ablations on the estimation problem for this dataset (c) predictive analysis through deep learning approaches and comparison with other arts and machine learning algorithms. Unlike previous approaches that rely on feature engineering or logical function deduction, our approach is fully data-driven and thus highly generic with better performance across different prediction tasks., Comment: 10 pages, 3 figures

The performance of text-to-speech (TTS) systems heavily depends on spectrogram to waveform generation, also known as the speech reconstruction phase. The time required for the same is known as synthesis delay. In this paper, an approach to reduce speech synthesis delay has been proposed. It aims to enhance the TTS systems for real-time applications such as digital assistants, mobile phones, embedded devices, etc. The proposed approach applies Fast Griffin Lim Algorithm (FGLA) instead Griffin Lim algorithm (GLA) as vocoder in the speech synthesis phase. GLA and FGLA are both iterative, but the convergence rate of FGLA is faster than GLA. The proposed approach is tested on LJSpeech, Blizzard and Tatoeba datasets and the results for FGLA are compared against GLA and neural Generative Adversarial Network (GAN) based vocoder. The performance is evaluated based on synthesis delay and speech quality. A 36.58% reduction in speech synthesis delay has been observed. The quality of the output speech has improved, which is advocated by higher Mean opinion scores (MOS) and faster convergence with FGLA as opposed to GLA., Comment: Accepted for publication in Springer Multimedia Tools and Applications Journal