S M Jishanul Islam

E-MedViTr: Enhanced Vision Transformers with Registers for Biomedical Image Classification

Authors: Sadia Ahmmed, Taimur Rahman, S M Jishanul Islam, Al-Momen Reyad, Sonjoy Dey, James Anthony Purification and Dewan Md. Farid

6th International Conference on Electrical Engineering and Information & Communication Technology 2024 [Scopus Indexed]

DOI: 10.1109/ICEEICT62016.2024.10534573

About: Medical image classification is critical in clinical decision-making, requiring efficient and precise evaluation of images. We propose an E-MedViTR model that boosts the performance of the vision transformer with registers by adding an enhanced classifier head. Research in this field could pave the way for more personalized, accurate, and efficient diagnostic procedures in medical imaging, benefiting both clinicians and patients.

MIMIC: Multimodal Islamophobic Meme Identification and Classification

Authors: S M Jishanul Islam, Sahid Hossain Mustakim, Sadia Ahmmed, Md. Faiyaz Abdullah Sayeedi, Swapnil Khondoker, Syed Tasdid Azam Dhrubo, and Nahid Hossain.

Muslims in ML Workshop, 38th Neural Information Processing Systems 2024 (NeurIPS 2024)

arXiv: arxiv.org/abs/2412.00681 || OpenReview || Code & Data

Abstract: Anti-Muslim hate speech has emerged within memes, characterized by context-dependent and rhetorical messages using text and images that seemingly mimic humor but convey Islamophobic sentiments. This work presents a novel dataset and proposes a classifier based on the Vision-and-Language Transformer (ViLT) specifically tailored to identify anti-Muslim hate within memes by integrating both visual and textual representations. Our model leverages joint modal embeddings between meme images and incorporated text to capture nuanced Islamophobic narratives that are unique to meme culture, providing both high detection accuracy and interoperability.

An audio video-based multi-modal fusion approach for emotion recognition

Authors: S M Jishanul Islam (FA), Sahid Hossain Mustakim, Musfirat Hossain, Mysun Mashira, Nur Islam Shourav, Md. Rayhan Ahmed (Corresponding Author), Salekul Islam (Corresponding Author), Swakkhar Shatabda, and A.K.M Muzahidul Islam

Under Review, Knowledge-based Systems (Elsevier)

About: A multi-modal fusion approach that takes audio and video input in parallel to effectively predict human emotions. Competes fairly with state-of-the-art models. Potential applications in the field of human-computer interaction (HCI), psychological evaluations, driver safety in vehicles, and so on.