K
Koushik Roy
Researcher at North South University
Publications - 9
Citations - 39
Koushik Roy is an academic researcher from North South University. The author has contributed to research in topics: Computer science & License. The author has an hindex of 1, co-authored 5 publications receiving 5 citations.
Papers
More filters
Proceedings ArticleDOI
A-DeepPixBis: Attentional Angular Margin for Face Anti-Spoofing
TL;DR: This study proposes a variant of BCE that enforces a margin in angular space and incorporate it in training the DeepPixBis model and presents a method to incorporate such a loss for attentive pixel wise supervision applicable in a fully convolutional setting.
Journal ArticleDOI
Bi-FPNFAS: Bi-Directional Feature Pyramid Network for Pixel-Wise Face Anti-Spoofing by Leveraging Fourier Spectra
Koushik Roy,Md. Al Mehedi Hasan,Labiba Rupty,Md. Sourave Hossain,Shirshajit Sengupta,Shehzad Noor Taus,Nabeel Mohammed +6 more
TL;DR: In this paper, a bi-directional feature pyramid network (BiFPN) was employed for convolutional multi-scaled feature extraction on the EfficientDet detection architecture, which is novel to the task of face anti-spoofing.
Journal ArticleDOI
BLPnet: A new DNN model and Bengali OCR engine for Automatic License Plate Recognition
Md. Saif Hassan Onim,Hussain Nyeem,Koushik Roy,Md. Mahmudul Hasan,Abtahi Ishmam,Md. Akiful Hoque Akif,Tareque Bashar Ovi +6 more
TL;DR: A computationally efficient and reasonably accurate Automatic License Plate Recognition (ALPR) system for Bengali characters with a new end-to-end DNN model that the model is characters rotation invariant, and can readily extract, detect and output the complete license plate number of a vehicle.
Proceedings ArticleDOI
A Robust Webcam-based Eye Gaze Estimation System for Human-Computer Interaction
Koushik Roy,Dibaloke Chanda +1 more
TL;DR: This research work approached gaze detection as a multiclass classification problem, this reduced the complexity of the method and allows the solution to be implemented in a way that allows free head movement of the user.
Proceedings ArticleDOI
An Analytical Approach for Enhancing the Automatic Detection and Recognition of Skewed Bangla License Plates
Koushik Roy,Abu Mohammad Shabbir Khan,Mohammad Ali,Sazid Rahman Simanto,Nabeel Mohammed,Muhammad Asif Atick,Shahidul Islam,Kazi Mejbaul Islam +7 more
TL;DR: This paper demonstrates how existing ALPR systems can be treated as black boxes and analyzed to understand what sort of license plate images they work best on and introduces a novel pipeline that combines deep learning and an algorithmic procedure for transforming images of both normal and skewed license plates into formats that are best suited for theALPR systems.