Children diagnosed with Attention-Deficit/Hyperactivity Disorder (ADHD) face many difficulties in maintaining their concentration (in terms of attention levels) and controlling their behaviors. Previous studies have mainly focused on identifying brain regions involved in cognitive processes or classifying ADHD and control subjects. However, the classification of attention levels of ADHD subjects has not yet been explored. Here, a robust Swin Transformer (Swin-T) model is proposed to classify the attention levels of ADHD subjects. The experimental cognitive task ‘Surround suppression’ includes two events: Stim ON and Stim OFF related to the high and low attention levels of a subject. In the proposed framework, ADHD-specific channels are initially identified from input Electroencephalography (EEG). Next, the significant, non-noisy connectivity features are extracted from those channels through the Singular Value Decomposition (SVD) method. Finally, the non-noisy features are passed to the robust Swin-T model for attention-level classification. The proposed model achieves 97.28% classification accuracy with 12 subjects. The robustness of the proposed model leads to potential benefits in EEG-based research and clinical settings, enhancing the reliability of ADHD assessments.

Perspective distortion (PD) leads to substantial alterations in the shape, size, orientation, angles, and spatial relationships of visual elements in images. Accurately determining camera intrinsic and extrinsic parameters is challenging, making it hard to synthesize perspective distortion effectively. The current distortion correction methods involve removing distortion and learning vision tasks, thus making it a multi-step process, often compromising performance. Recent work leverages the Möbius transform for mitigating perspective distortions (MPD) to synthesize perspective distortions without estimating camera parameters. Möbius transform requires tuning multiple interdependent and interrelated parameters and involving complex arithmetic operations, leading to substantial computational complexity. To address these challenges, we propose Log Conformal Maps (LCM), a method leveraging the logarithmic function to approximate perspective distortions with fewer parameters and reduced computational complexity. We provide a detailed foundation complemented with experiments to demonstrate that LCM with fewer parameters approximates the MPD. We show that LCM integrates well with supervised and self-supervised representation learning, outperform standard models, and matches the state-of-the-art performance in mitigating perspective distortion over multiple benchmarks, namely Imagenet-PD, Imagenet-E, and Imagenet-X. Further LCM demonstrate seamless integration with person re-identification and improved the performance. Source code is made publicly available at https://github.com/meenakshi23/Log-Conformal-Maps. 

Perspective distortion (PD) causes unprecedented changes in shape, size, orientation, angles, and other spatial relationships of visual concepts in images. Precisely estimating camera intrinsic and extrinsic parameters is a challenging task that prevents synthesizing perspective distortion. Non-availability of dedicated training data poses a critical barrier to developing robust computer vision methods. Additionally, distortion correction methods make other computer vision tasks a multi-step approach and lack performance. In this work, we propose mitigating perspective distortion (MPD) by employing a fine-grained parameter control on a specific family of Möbius transform to model real-world distortion without estimating camera intrinsic and extrinsic parameters and without the need for actual distorted data. Also, we present a dedicated perspectively distorted benchmark dataset, ImageNet-PD, to benchmark the robustness of deep learning models against this new dataset. The proposed method outperforms existing benchmarks, ImageNet-E and ImageNet-X. Additionally, it significantly improves performance on ImageNet-PD while consistently performing on standard data distribution. Notably, our method shows improved performance on three PD-affected real-world applications—crowd counting, fisheye image recognition, and person re-identification—and one PD-affected challenging CV task: object detection. The source code, dataset, and models are available on the project webpage at https://prakashchhipa.github.io/projects/mpd.

This paper addresses the critical challenge of vehicle detection in the harsh winter conditions in the Nordic regions, characterized by heavy snowfall, reduced visibility, and low lighting. Due to their susceptibility to environmental distortions and occlusions, traditional vehicle detection methods have struggled in these adverse conditions. The advanced proposed deep learning architectures brought promise, yet the unique difficulties of detecting vehicles in Nordic winters remain inadequately addressed. This study uses the Nordic Vehicle Dataset (NVD), which contains UAV (unmanned aerial vehicle) images from northern Sweden, to evaluate the performance of state-of-the-art vehicle detection algorithms under challenging weather conditions. Our methodology includes a comprehensive evaluation of single-stage, two-stage, segmentation-based, and transformer-based detectors against the NVD. We propose a series of enhancements tailored to each detection framework, including data augmentation, hyperparameter tuning, transfer learning, and Specifically implementing and enhancing the Detection Transformer (DETR). A novel architecture is proposed that leverages self-attention mechanisms with the help of MSER (maximally stable extremal regions) and RST (Rough Set Theory) to identify and filter the region that model long-range dependencies and complex scene contexts. Our findings not only highlight the limitations of current detection systems in the Nordic environment but also offer promising directions for enhancing these algorithms for improved robustness and accuracy in vehicle detection amidst the complexities of winter landscapes. The code and the dataset are available at https://nvd.ltu-ai.dev.

Scene text image super-resolution (STISR) aims to enhance the resolution of text images while simultaneously improving their readability by reducing noise, blur, and other degradations. Existing diffusion-based approaches for STISR primarily rely on text-prior information but often overlook the importance of explicitly modeling the visual structure of the text. In this paper, we propose a novel Skeleton-Aware Diffusion Method (SADM) for STISR, which introduces text skeletons as structural guidance to the diffusion process. The text skeleton serves as a critical visual cue, helping the model to better restore the fine details of text, even in severely degraded low-resolution images. Generating high-quality skeletons from low-resolution scene text is a challenging task due to the inherent blurring and noise present in such images. To tackle this, we introduce a diffusion-based Skeleton Correction Network (SCN), which refines the initial skeletons produced by a convolutional neural network-based skeletonization model. The SCN effectively improves the accuracy of the skeletons, allowing for more precise structural guidance during the diffusion process. Our extensive experiments demonstrate the significant benefits of incorporating skeleton information into the STISR pipeline. The proposed SADM achieves state-of-the-art performance on the TextZoom dataset, with accuracies of 81.4%, 64.9%, and 49.6% on the easy, medium, and hard subsets, respectively, compared to the previous best results by ASTER text recognizer. Through detailed analysis, we also show that improving the quality of skeletons from low-resolution images leads to better super-resolution outcomes and enhances the performance of text recognizers.

In this paper, we introduce Fast&Focused-Net (FFN), a novel deep neural network architecture tailored for efficiently encoding small objects into fixed-length feature vectors. Contrary to conventional Convolutional Neural Networks, FFN employs a series of our newly proposed layer, the Volume-wise Dot Product (VDP) layer, designed to address several inherent limitations of CNNs. Specifically, CNNs exhibit a smaller effective receptive field (ERF) than their theoretical counterparts, limiting their vision span. Additionally, the initial layers in CNNs produce low-dimensional feature vectors, presenting a bottleneck for subsequent learning. Lastly, the computational overhead of CNNs, particularly in capturing diverse image regions by parameter sharing, is significantly high. The VDP layer, at the heart of FFN, aims to remedy these issues by efficiently covering the entire image patch information with reduced computational. Experimental results demonstrate the prowess of FFN in a variety of applications. Our network outperformed state-of-the-art methods for small object classification tasks on datasets such as CIFAR-10, CIFAR-100, STL-10, SVHN-Cropped, and Fashion-MNIST. In the context of larger image classification, when combined with a transformer encoder (ViT), FFN produced competitive results for OpenImages V6, ImageNet-1K, and Places365 datasets. Moreover, the same combination showcased unparalleled performance in text recognition tasks across SVT, IC15, SVTP, and HOST datasets.

Model sparsification in deep learning promotes simpler, more interpretable models with fewer parameters. This not only reduces the model’s memory footprint and computational needs but also shortens inference time. This work focuses on creating sparse models optimized for multiple tasks with fewer parameters. These parsimonious models also possess the potential to match or outperform dense models in terms of performance. In this work, we introduce channel-wise l1/l2 group sparsity in the shared convolutional layers parameters (or weights) of the multi-task learning model. This approach facilitates the removal of extraneous groups i.e., channels (due to l1 regularization) and also imposes a penalty on the weights, further enhancing the learning efficiency for all tasks (due to l2 regularization). We analyzed the results of group sparsity in both single-task and multi-task settings on two widely-used multi-task learning datasets: NYU-v2 and CelebAMask-HQ. On both datasets, which consist of three different computer vision tasks each, multi-task models with approximately 70% sparsity outperform their dense equivalents. We also investigate how changing the degree of sparsification influences the model’s performance, the overall sparsity percentage, the patterns of sparsity, and the inference time.

Despite their vulnerability to competent forgers, signatures are one of the most widely used user verification methods. Recent research has revealed that EEG signals are harder to reproduce and give superior biometric information. This study aims to improve the effectiveness of person authentication by using deep learning techniques on electroencephalogram (EEG) signals. The broad implementation of EEG-based authentication systems has been hindered by problems such as noise, variability, and inter-subject variances despite the potential distinctiveness of EEG signals. We propose a multiscale convolutional neural network (CNN) and a Bidirectional LSTM (BiLSTM) model called CNN-BiLSTM to extract features and classify raw EEG data. This methodology involves acquiring raw EEG data, preprocessing for noise reduction, standardization, normalization, and employing deep learning techniques for feature extraction and classification. Experimental results exhibit a notable improvement in accuracy and reliability compared to existing EEG authentication methods such as LOF, CNN, FCN, EfficientNet-B0, and BiLSTM. The results showcase the performance of the proposed deep learning model utilizing established metrics such as precision, sensitivity, specificity, and accuracy. The proposed methodology outperforms existing methods and achieves a training and validation accuracy of 98.9% and 92.2%, respectively. The findings of the research demonstrate that the proposed approach has been successful in achieving highly effective results by using EEG signals for the purpose of resolving issues related to person identification.

Integrating knowledge across different domains is an essential feature of human learning. Learning paradigms such as transfer learning, meta-learning, and multi-task learning reflect the human learning process by exploiting the prior knowledge for new tasks, encouraging faster learning and good generalization for new tasks. This article gives a detailed view of these learning paradigms and their comparative analysis. The weakness of one learning algorithm turns out to be a strength of another, and thus, merging them is a prevalent trait in the literature. Numerous research papers focus on each of these learning paradigms separately and provide a comprehensive overview of them. However, this article reviews research studies that combine (two of) these learning algorithms. This survey describes how these techniques are combined to solve problems in many different fields of research, including computer vision, natural language processing, hyper-spectral imaging, and many more, in a supervised setting only. Based on the knowledge accumulated from the literature, we hypothesize a generic task-agnostic and model-agnostic learning network – an ensemble of meta-learning, transfer learning, and multi-task learning, termed Multi-modal Multi-task Meta Transfer Learning. We also present some open research questions, limitations, and future research directions for this proposed network. The aim of this article is to spark interest among scholars in effectively merging existing learning algorithms with the intention of advancing research in this field. Instead of presenting experimental results, we invite readers to explore and contemplate techniques for merging algorithms while navigating through their limitations.