Abdul Muntakim Rafi

For the deaf and dumb (D&D) people, sign language is one of the primary and most used methods for communication. All over the world, every day the D&D community faces difficulties while communicating with the general mass. Most of the times, they need an interpreter to communicate with others and the interpreter may not always be available. The issue is also faced by the people using Bengali Sign Language (BdSL) due to the lack of BdSL interpreters. Recently, computer vision-based systems have been introduced for automatic recognition of sign languages to mitigate this problem. But so far the number of reliable works done for the recognition of BdSL is not adequate. In this paper, we propose a method for automatic detection of BdSL alphabets. Our system solely relies on the images of bare hands, which allows the users to interact with the system in a natural way. We have collected in total 12581 different hand signs for the 38 BdSL alphabets in collaboration with the National Federation of the Deaf. We propose a VGG19 based convolutional neural network for the recognition of 38 classes and achieve an overall test accuracy of 89.6%.

MOTIVATION: The increasing volume of data from high-throughput experiments including parallel reporter assays facilitates the development of complex deep-learning approaches for modeling DNA regulatory grammar. RESULTS: Here, we introduce LegNet, an EfficientNetV2-inspired convolutional network for modeling short gene regulatory regions. By approaching the sequence-to-expression regression problem as a soft classification task, LegNet secured first place for the autosome.org team in the DREAM 2022 challenge of predicting gene expression from gigantic parallel reporter assays. Using published data, here, we demonstrate that LegNet outperforms existing models and accurately predicts gene expression per se as well as the effects of single-nucleotide variants. Furthermore, we show how LegNet can be used in a diffusion network manner for the rational design of promoter sequences yielding the desired expression level. AVAILABILITY AND IMPLEMENTATION: https://github.com/autosome-ru/LegNet. The GitHub repository includes Jupyter Notebook tutorials and Python scripts under the MIT license to reproduce the results presented in the study.

Camera model identification has earned paramount importance in the field of image forensics with an upsurge of digitally altered images which are constantly being shared through websites, media, and social applications. But, the task of identification becomes quite challenging if metadata are absent from the image and/or if the image has been post-processed. In this paper, we present a DenseNet pipeline to solve the problem of identifying the source camera-model of an image. Our approach is to extract patches of 256*256 from a labeled image dataset and apply augmentations, i.e., Empirical Mode Decomposition (EMD). We use this extended dataset to train a Neural Network with the DenseNet-201 architecture. We concatenate the output features for 3 different sizes (64*64, 128*128, 256*256) and pass them to a secondary network to make the final prediction. This strategy proves to be very robust for identifying the source camera model, even when the original image is post-processed. Our model has been trained and tested on the Forensic Camera-Model Identification Dataset provided for the IEEE Signal Processing (SP) Cup 2018. During testing we achieved an overall accuracy of 98.37%, which is the current state-of-the-art on this dataset using a single model. We used transfer learning and tested our model on the Dresden Database for Camera Model Identification, with an overall test accuracy of over 99% for 19 models. In addition, we demonstrate that the proposed pipeline is suitable for other image-forensic classification tasks, such as, detecting the type of post-processing applied to an image with an accuracy of 96.66% -- which indicates the generality of our approach.

A systematic evaluation of how model architectures and training strategies impact genomics model performance is needed. To address this gap, we held a DREAM Challenge where competitors trained models on a dataset of millions of random promoter DNA sequences and corresponding expression levels, experimentally determined in yeast. For a robust evaluation of the models, we designed a comprehensive suite of benchmarks encompassing various sequence types. All top-performing models used neural networks but diverged in architectures and training strategies. To dissect how architectural and training choices impact performance, we developed the Prix Fixe framework to divide models into modular building blocks. We tested all possible combinations for the top three models, further improving their performance. The DREAM Challenge models not only achieved state-of-the-art results on our comprehensive yeast dataset but also consistently surpassed existing benchmarks on Drosophila and human genomic datasets, demonstrating the progress that can be driven by gold-standard genomics datasets.