Fisseha A. Ferede

I am a PhD student at The University of Memphis where I am advised by Dr. Balasubramanian. My current research area lies at the intersection of computer vision and machine learning, particularly in optical flow estimation, video frame interpolation and future frame prediction, and video compression problems.

As a computer vision and ML research intern at St. Jude Children's Research Hospital, I developed and deployed AI-driven software solutions for isotropic reconstruction of 3D medical images. My work focused on developing optical flow-guided reconstruction of 3D volumes to enhance the axial resolution of medical images, which is often lower than the lateral resolution due to the physics of the imaging system.

Previously, I was a research intern at BIESL, KAIST where I developed AI-enabled programmable and wearable devices for health monitoring applications. I was also a summer research intern at MST lab, TU Chemnitz where I designed and built 3D mechanical components and electrical system of a gesture controlled robotic arm.

I received my MS degree in Computer Engineering (Computer Vision focus) from University of Memphis and my BS degree in Electrical Engineering from KAIST.

Email  /  Google Scholar  /  Linkedin  /  Github

profile photo
Research
SSTM: Spatiotemporal Recurrent Transformers for Multi-frame Optical Flow Estimation
Fisseha A. Ferede, Madhusudhanan Balasubramanian
Neurocomputing, 2023
arXiv / poster

Multi-frame based optical flow estimation algorithm.

KinemaNet: Kinematic descriptors of deformation of ONH images for glaucoma progression detection
Fisseha A. Ferede, Madhusudhanan Balasubramanian
In preparation for MIA submission
project page / code / software

Novel structural biomarkers for glaucoma progression detection.

Z-upscaling: Optical Flow Guided Frame Interpolation for Isotropic Reconstruction of 3D EM Volumes
Fisseha A. Ferede, Ali Khalighifar, Jaison John, Krishnan Venkataraman, Khaled Khairy
IEEE International Symposium on Biomedical Imaging (ISBI) , 2025
arXiv / code

Spatial flow interpolation approach for isotropic reconstruction of 3D medical volumes.

Certifications
Academic Services
  • Journal Reviewer: Scientific Reports-Nature
  • Teaching Assistant: Software Engineering, Signals and Systems at The University of Memphis
Honors and Awards

Visitor Map


Template borrowed from Jon Barron's website.