Hi, I'm Sayeed

I am a Graduate Research Assistant under the supervision of Professor Saibal Mukhopadhyay at the Green Lab at Georgia Institute of Technology, Atlanta. I have served in the position of lecturer, Department of Computer Science and Engineering at BRAC University from June 2022 to July 2023 and am currently on a study leave. I completed my B.Sc. (2022) from the Department of Electrical and Electronic Engineering, Bangladesh University of Engineering and Technology.

I am currently working on the closed loop control of quadrupedal and aerial robots based on vision, LIDAR and event camera. I am highly interested in research in the fields of perception and planning of autonomous mobile, aerial, bipedal and quadrupedal robots, and robotic manipulators. Being an avid fan of robotics and automation, I enjoy designing automatic machines and watching them come to life. I believe that the unification of robotics and AI is the path toward a new future. During my undergraduate studies, I completed several projects related to electrical engineering, robotics, and computer vision. I have also conducted several workshops related to embedded systems and the basics of robotics.

Mir Sayeed Mohammad

Graduate Research Assistant, Georgia Tech

My CV

Education

(2025) - M.Sc in Electrical and Computer Engineering - Georgia Institute of Technology
(2022) - B.Sc in Electrical and Electronics Engineering - Bangladesh University of Engineering and Technology (BUET)
(2016) - H.S.C - Notre Dame College, Dhaka
(2014) - S.S.C - Engineering University School and College, Dhaka

Research Interests

Robotic Perception and Planning
Autonomous Vehicles
Computer Vision and Image Processing
Robotic Manipulation
Autonomous Mobile Robots
Legged Robots
Robot Learning
Deep Learning and Reinforcement Learning
Biomedical Image and Signal Processing

News

2025 September Our paper SSMRadNet accepted for publication at WACV2025.
2025 May I got my Master's degree in ECE from Georgia Tech.
2025 April Our paper Adaptive-Cloud accepted for publication at IEEE Robotics and Automation Letters and will be presented at ICRA2026.
2024 October My work on Adaptive-Cloud was selected as the first runner's up in Cognisense Annual Review 2024 student presentations.
2023 August I started my PhD program as a graduate research assistant in ECE, Georgia Tech under professor Saibal's lab.
2023 July Left my job at BRAC University on a study leave.
2022 June Started job as a lecturer in CSE at BRAC University.
2022 May Completed my undergrad in EEE at Bangladesh University of Engineering and Technology.

Publication Highlights

SSMRadNet : A Sample-wise State-Space Framework for Efficient and Ultra-Light Radar Segmentation and Object Detection.

Anuvab Sen, Mir Sayeed Mohammad, Saibal Mukhopadhyay

WACV, 2026

We introduce SSMRadNet, the first multi-scale State Space Model (SSM) based detector for Frequency Modulated Continuous Wave (FMCW) radar that sequentially processes raw ADC samples through two SSMs. One SSM learns a chirp-wise feature by sequentially processing samples from all receiver channels within one chirp, and a second SSM learns a representation of a frame by sequentially processing chirp-wise features. The latent representations of a radar frame are decoded to perform segmentation and detection tasks. Comprehensive evaluations on the RADIal dataset show SSMRadNet has 10-33× fewer parameters and 60-88× less computation (GFLOPs) while being 3.7× faster than state-of-the-art transformer and convolution-based radar detectors at competitive performance for segmentation tasks.

Adaptive-Cloud: Dynamic Computation Control for 3D Object Detection From LIDAR Point Clouds.

Mir Sayeed Mohammad, Uday Kamal, Saibal Mukhopadhyay

IEEE Robotics and Automation Letters, 2025

In this work, we introduce an adaptive hierarchical framework for efficient 3D object detection from point cloud data, designed to dynamically balance computational efficiency and detection performance. Our approach employs a shared feature extractor and multiple detector backbones of varying widths, enabling selective activation of models based on the complexity of the input scene. A novel feature gating mechanism dynamically determines the most relevant features for reduced-width backbones, while a surrogate loss prediction module ranks models in real-time, ensuring optimal backbone selection with minimal overhead. This adaptive strategy reduces compute costs by 41.4% while maintaining a negligible 2.44% reduction in detection accuracy across a range of real-world driving scenes (urban, highway, residential, campus, person) from the KITTI dataset. By addressing runtime adaptability—a critical gap in existing 3D detection frameworks—our method provides a significant algorithmic improvement for high-performance detection models in resource-constrained environments.

S2C-DeLeNet: A Parameter Transfer Based Segmentation-Classification Integration for Detecting Skin Cancer Lesions from Dermoscopic Images.

Md. Jahin Alam, Mir Sayeed Mohammad, Md Adnan Faisal Hossain, Ishtiaque Ahmed Showmik, Munshi Sanowar Raihan, Shahed Ahmed, Talha Ibn Mahmud

Computers in Biology and Medicine, Elsevier, 2022

Dermoscopic images ideally depict pigmentation attributes on the skin surface which is highly regarded in the medical community for detection of skin abnormality, disease or even cancer. The identification of such abnormality, however, requires trained eyes and accurate detection necessitates the process being time-intensive. As such, computerized detection schemes have become quite an essential, especially schemes which adopt deep learning tactics. In this paper, a convolutional deep neural network, S2C-DeLeNet, is proposed, which (i) Performs segmentation procedure of lesion based regions with respect to the unaffected skin tissue from dermoscopic images using a segmentation sub-network, (ii) Classifies each image based on its medical condition type utilizing transferred parameters from the inherent segmentation sub-network. The architecture of the segmentation sub-network contains EfficientNet-B4 backbone in place of the encoder and the classification sub-network bears a ‘Classification Feature Extraction’ system which pulls trained segmentation feature maps towards lesion prediction. Inside the classification architecture, there have been designed, (i) A ‘Feature Coalescing Module’ in order to trail and mix each dimensional feature from both encoder and decoder, (ii) A ‘3D-Layer Residuals’ block to create a parallel pathway of low-dimensional features with high variance for better classification. After fine-tuning on a publicly accessible dataset, a mean dice-score of 0.9494 during segmentation is procured which beats existing segmentation strategies and a mean accuracy of 0.9103 is obtained for classification which outperforms conventional and noted classifiers. Additionally, the already fine-tuned network demonstrates highly satisfactory results on other skin cancer segmentation datasets while cross-inference. Extensive experimentation is done to prove the efficacy of the network for not only dermoscopic images but also different medical modalities; which can show its potential in being a systematic diagnostic solution in the field of dermatology and possibly more.

A Seuquence Agnostic Multimodal Preprocessing for Clogged Blood Vessel Detection in Alzheimer's Diagnosis.

Partha Ghosh, Md Abrar Istiak, Mir Sayeed Mohammad, Swapnil Saha, Uday Kamal

IEEE ICASSPW, 2023

Successful identification of blood vessel blockage is a crucial step for Alzheimer’s disease diagnosis. These blocks can be identified from the spatial and time-depth variable Two-Photon Excitation Microscopy (TPEF) images of the brain blood vessels using machine learning methods. In this study, we propose several pre-processing schemes to improve the performance of these methods. Our method includes 3D-point cloud data extraction from image modality and their feature-space fusion to leverage complementary information inherent in different modalities. We also enforce the learned representation to be sequence-order invariant by utilizing bidirection dataflow. Experimental results on The Clog Loss dataset1 show that our proposed method consistently outperforms the state-of-the-art pre-processing methods in stalled and non-stalled vessel classification.