Akbar Ali
I am a fourth-year Ph.D. student in Computer Science and Engineering at IIT Gandhinagar, advised by Prof. Shanmuganathan Raman. My research is on generative priors for image restoration — latent inversion in StyleGAN and diffusion models, zero-shot and training-free restoration, and 2D/3D Gaussian splatting for hyperspectral imaging. I am currently a visiting researcher at HZDR in Germany through the Helmholtz Institute Freiberg research-abroad program.
Before the Ph.D. I was a project associate at IIT Ropar, working on volume estimation of complex objects from hyperspectral and stereo imaging. I am also a licensed UAV pilot and have run drone campaigns for agricultural and geological data collection.
News
- [August 2026] Submitted a paper to WACV 2027.
- [July 2026] Submitted two papers to ACCV 2026.
- [April 2026] GenR accepted to Pattern Recognition Letters. [paper] [code]
- [March 2026] Started as a visiting researcher at HZDR, Germany, through the Helmholtz Institute Freiberg research-abroad program.
- [September 2025] Presented two papers at ICIP 2025 in Anchorage, USA — my first international conference.
- [June 2025] COT-AD and IRUSIR both accepted to ICIP 2025.
- [January 2025] Lead Teaching Assistant for ES 113, Data Centric Computing.
- [November 2024] Third drone and core acquisition campaign with the Earth Science group.
- [August 2024] Graduate Teaching Fellow for FP 602, Writing.
- [July 2024] Transport evaluation field visits to Silvassa and Aurangabad.
- [June 2024] Attended the 3D Vision Summer School at IIIT Bangalore.
- [June 2024] Teaching Assistant for SC 333, a short course on drone data acquisition and processing.
- [May 2024] Attended the Visvesvaraya PhD Scheme workshop at IIT Mandi.
- [February 2024] Started work on 3D geophysical image translation with GANs.
- [June 2023] Began the cotton smart-farming campaign — weekly drone missions across a full growth cycle.
- [May 2023] First field campaign with the Earth Science group.
- [Early 2023] Trained on drone operation at IIT Gandhinagar and certified as a UAV pilot for platforms up to 25 kg.
- [February 2023] Parkinson's disease prediction chapter published by Springer Nature.
- [December 2022] Joined IIT Gandhinagar as a Ph.D. scholar.
- [2022] Qualified UGC-NET in Computer Applications.
- [September 2021] Joined IIT Ropar as a project associate.
- [July 2021] Completed M.Tech at Aliah University, second rank overall.
- [July 2019] Completed B.Tech at Aliah University, first rank overall.
Research
A degradation-agnostic face restoration method that optimises directly in the latent space of a pretrained generator, so one model handles denoising, deblurring and inpainting without retraining per degradation.
Pivotal tuning inversion anchors a blurred observation to a generator's latent manifold, recovering sharp detail without paired supervision or a known blur kernel.
Denoising a single image with no training and no clean reference, using CLIP as a semantic critic to steer restoration towards a perceptually faithful result.
A dataset and recognition pipeline for historical stone inscriptions, where weathering, low contrast and archaic glyph forms break standard scene-text OCR.
A stress test of unsupervised StyleGAN restoration under degradations it was never tuned for, showing where latent-space optimisation holds up and where it silently fails.
An aerial and ground image dataset covering a full cotton growth cycle, collected over weekly drone missions at 10 m, 15 m and 115 m from July to December 2023, for crop health and pest analysis.
A predictive framework for Parkinson's disease analysis pairing classification accuracy with descriptive feature analysis, so clinicians can see which measurements drive each prediction.
Education
Volume estimation of complex objects using hyperspectral imaging and stereo vision fusion.
Teaching
Mentored 45 students in data structures, algorithms, data analysis in Python (Pandas, Matplotlib), SQL and introductory machine learning. Ran QA sessions and coordinated grading with the instructor.
Mentored 35 students through classwork, exams and academic progress tracking.
Short course for ~30 students on flight training, data acquisition and processing with RGB and multispectral imagery, including hands-on fieldwork.
Ran labs and tutorials for 50+ students using the Python data science stack; also assisted the Probability and Randomization course.
Mentored B.Tech and M.Tech students on hyperspectral imaging, image editing and GAN inversion projects.
Projects and Field Work
Translated 3D geophysical scans into photorealistic virtual outcrop geology, with a method for generating diverse outcrop appearances at varying altitudes.
Studied the effect of weather and pests on crop health across a full growth cycle, flying weekly drone missions at 10 m, 15 m and 115 m from July to December 2023.
Three field campaigns (May 2023, Aug 2024, Nov 2024) for aerial and core sample collection. Site names withheld for confidentiality.
Data collection visits to Silvassa and Aurangabad.
Week-long school on deep learning for 3D vision: lectures, hands-on sessions and research discussions.
Workshop on research presentation and interdisciplinary collaboration.
Awards and Certifications
DJI Mavic and Bajrang platforms. Mission planning, pre-flight checks, spectral data capture and aerial inspection.
Three during B.Tech and two during M.Tech.