Mahmudul Hasan

Md Mohaiminul Islam, Mahmudul Hasan, Kishan Shamsundar Athrey et al.

The ability to distinguish between different movie scenes is critical for understanding the storyline of a movie. However, accurately detecting movie scenes is often challenging as it requires the ability to reason over very long movie segments. This is in contrast to most existing video recognition models, which are typically designed for short-range video analysis. This work proposes a State-Space Transformer model that can efficiently capture dependencies in long movie videos for accurate movie scene detection. Our model, dubbed TranS4mer, is built using a novel S4A building block, which combines the strengths of structured state-space sequence (S4) and self-attention (A) layers. Given a sequence of frames divided into movie shots (uninterrupted periods where the camera position does not change), the S4A block first applies self-attention to capture short-range intra-shot dependencies. Afterward, the state-space operation in the S4A block is used to aggregate long-range inter-shot cues. The final TranS4mer model, which can be trained end-to-end, is obtained by stacking the S4A blocks one after the other multiple times. Our proposed TranS4mer outperforms all prior methods in three movie scene detection datasets, including MovieNet, BBC, and OVSD, while also being $2\times$ faster and requiring $3\times$ less GPU memory than standard Transformer models. We will release our code and models.

Tahmida Mahmud, Mohammad Billah, Mahmudul Hasan et al.

Most of the existing works on human activity analysis focus on recognition or early recognition of the activity labels from complete or partial observations. Similarly, almost all of the existing video captioning approaches focus on the observed events in videos. Predicting the labels and the captions of future activities where no frames of the predicted activities have been observed is a challenging problem, with important applications that require anticipatory response. In this work, we propose a system that can infer the labels and the captions of a sequence of future activities. Our proposed network for label prediction of a future activity sequence has three branches where the first branch takes visual features from the objects present in the scene, the second branch takes observed sequential activity features, and the third branch captures the last observed activity features. The predicted labels and the observed scene context are then mapped to meaningful captions using a sequence-to-sequence learning-based method. Experiments on four challenging activity analysis datasets and a video description dataset demonstrate that our label prediction approach achieves comparable performance with the state-of-the-arts and our captioning framework outperform the state-of-the-arts.

Mahmudul Hasan, Sujoy Paul, Anastasios I. Mourikis et al.

Activity recognition is a challenging problem with many practical applications. In addition to the visual features, recent approaches have benefited from the use of context, e.g., inter-relationships among the activities and objects. However, these approaches require data to be labeled, entirely available beforehand, and not designed to be updated continuously, which make them unsuitable for surveillance applications. In contrast, we propose a continuous-learning framework for context-aware activity recognition from unlabeled video, which has two distinct advantages over existing methods. First, it employs a novel active-learning technique that not only exploits the informativeness of the individual activities but also utilizes their contextual information during query selection; this leads to significant reduction in expensive manual annotation effort. Second, the learned models can be adapted online as more data is available. We formulate a conditional random field model that encodes the context and devise an information-theoretic approach that utilizes entropy and mutual information of the nodes to compute the set of most informative queries, which are labeled by a human. These labels are combined with graphical inference techniques for incremental updates. We provide a theoretical formulation of the active learning framework with an analytic solution. Experiments on six challenging datasets demonstrate that our framework achieves superior performance with significantly less manual labeling.

Md Zahangir Alom, Mahmudul Hasan, Chris Yakopcic et al.

Machine learning and computer vision have driven many of the greatest advances in the modeling of Deep Convolutional Neural Networks (DCNNs). Nowadays, most of the research has been focused on improving recognition accuracy with better DCNN models and learning approaches. The recurrent convolutional approach is not applied very much, other than in a few DCNN architectures. On the other hand, Inception-v4 and Residual networks have promptly become popular among computer the vision community. In this paper, we introduce a new DCNN model called the Inception Recurrent Residual Convolutional Neural Network (IRRCNN), which utilizes the power of the Recurrent Convolutional Neural Network (RCNN), the Inception network, and the Residual network. This approach improves the recognition accuracy of the Inception-residual network with same number of network parameters. In addition, this proposed architecture generalizes the Inception network, the RCNN, and the Residual network with significantly improved training accuracy. We have empirically evaluated the performance of the IRRCNN model on different benchmarks including CIFAR-10, CIFAR-100, TinyImageNet-200, and CU3D-100. The experimental results show higher recognition accuracy against most of the popular DCNN models including the RCNN. We have also investigated the performance of the IRRCNN approach against the Equivalent Inception Network (EIN) and the Equivalent Inception Residual Network (EIRN) counterpart on the CIFAR-100 dataset. We report around 4.53%, 4.49% and 3.56% improvement in classification accuracy compared with the RCNN, EIN, and EIRN on the CIFAR-100 dataset respectively. Furthermore, the experiment has been conducted on the TinyImageNet-200 and CU3D-100 datasets where the IRRCNN provides better testing accuracy compared to the Inception Recurrent CNN (IRCNN), the EIN, and the EIRN.

Md Zahangir Alom, Peheding Sidike, Mahmudul Hasan et al.

In spite of advances in object recognition technology, Handwritten Bangla Character Recognition (HBCR) remains largely unsolved due to the presence of many ambiguous handwritten characters and excessively cursive Bangla handwritings. Even the best existing recognizers do not lead to satisfactory performance for practical applications related to Bangla character recognition and have much lower performance than those developed for English alpha-numeric characters. To improve the performance of HBCR, we herein present the application of the state-of-the-art Deep Convolutional Neural Networks (DCNN) including VGG Network, All Convolution Network (All-Conv Net), Network in Network (NiN), Residual Network, FractalNet, and DenseNet for HBCR. The deep learning approaches have the advantage of extracting and using feature information, improving the recognition of 2D shapes with a high degree of invariance to translation, scaling and other distortions. We systematically evaluated the performance of DCNN models on publicly available Bangla handwritten character dataset called CMATERdb and achieved the superior recognition accuracy when using DCNN models. This improvement would help in building an automatic HBCR system for practical applications.

Mahmudul Hasan, Jonghyun Choi, Jan Neumann et al.

Perceiving meaningful activities in a long video sequence is a challenging problem due to ambiguous definition of 'meaningfulness' as well as clutters in the scene. We approach this problem by learning a generative model for regular motion patterns, termed as regularity, using multiple sources with very limited supervision. Specifically, we propose two methods that are built upon the autoencoders for their ability to work with little to no supervision. We first leverage the conventional handcrafted spatio-temporal local features and learn a fully connected autoencoder on them. Second, we build a fully convolutional feed-forward autoencoder to learn both the local features and the classifiers as an end-to-end learning framework. Our model can capture the regularities from multiple datasets. We evaluate our methods in both qualitative and quantitative ways - showing the learned regularity of videos in various aspects and demonstrating competitive performance on anomaly detection datasets as an application.