Toqi Tahamid Sarker

Toqi Tahamid Sarker

I study computer vision — with a focus on deep learning for precision agriculture, environmental monitoring, and food safety. Most of my work uses transformer-based architectures and multi-task learning on large-scale imagery, spanning livestock methane emissions, rumen health, crop phenotyping, weed management, and non-destructive food inspection. More recently, I've been working on vision-language models for open-vocabulary segmentation.

I'm a PhD candidate at Southern Illinois University Carbondale, advised by Dr. Khaled Ahmed at the BASE Lab. Before SIUC, I was a Graduate Research Assistant at Georgia State University's Panacea Lab, where I worked on solar event tracking, and a Software Engineer at Penguin Ecommerce in Dhaka. I'm on the 2026–27 academic and industry job market.

Open to 2026–27 rolesView Resume
Contacttoqitahamid.sarker [at] siu [dot] edu
Conference2026

FryNet: Dual-Stream Adversarial Fusion for Non-Destructive Frying Oil Oxidation Assessment

Khaled R. Ahmed, Toqi Tahamid Sarker, Taminul Islam, Tamany M. Alanezi, and Amer AbuGhazaleh

CVPR WorkshopsOral Presentation
Conference2026

WeedRepFormer: Reparameterizable Vision Transformers for Real-Time Waterhemp Segmentation and Gender Classification

T. T. Sarker, T. Islam, K. R. Ahmed, C. B. Rankrape, K. E. Creager, and K. Gage

arXiv preprint arXiv:2601.03431
Conference2025

GasTwinFormer: A Hybrid Vision Transformer for Livestock Methane Emission Segmentation and Dietary Classification in Optical Gas Imaging

T. T. Sarker, M. G. Embaby, T. Islam, A. AbuGhazaleh, and K. Ahmed

ICCV WorkshopOral PresentationBest Paper Nomination
Conference2025

WeedSense: Multi-Task Learning for Weed Segmentation, Height Estimation, and Growth Stage Classification

T. T. Sarker, K. Ahmed, T. Islam, C. B. Rankrape, and K. Gage

ICCV WorkshopOral Presentation
Conference2025

SoyStageNet: Balancing Accuracy and Efficiency for Real-Time Soybean Growth Stage Detection

A. Lakhssassi, T. T. Sarker, K. Ahmed, N. Lakhssassi, and K. Meksem

CVPR Workshop

News

June 2026
"Mask-Guided Multi-Task Learning: Real-Time RGB Prediction of Plant Photosynthetic Efficiency at the Edge" accepted at the International Conference on Pattern Recognition (ICPR) 2026.
June 2026
"WeedRepFormer: Reparameterizable Vision Transformers for Real-Time Waterhemp Segmentation and Gender Classification" accepted at the International Conference on Pattern Recognition (ICPR) 2026.
April 2026
One paper FryNet is accepted by CVPR Workshops 2026.
April 2026
Awarded Dissertation Research Assistantship for the 2026-2027 academic year.