Akbar Ali

My first international conference. Anchorage in September is something else — we finished sessions and still had daylight to wander.

Met people whose papers I had been reading for two years. That part never quite stops feeling strange.

Presenting COT-AD, the cotton dataset — the payoff for six months of weekly drone flights over a single field.

Presenting IRUSIR on the robustness of unsupervised StyleGAN restoration, the second of the two ICIP papers.

A week of lectures and hands-on sessions on deep learning for 3D vision. This is where Gaussian splatting stopped being a paper I had skimmed and started being something I wanted to build on.

Long days, good company. Half the value of a summer school is the conversations between sessions.

Presenting to an interdisciplinary room in the Himalayas. Explaining your work to people outside your field is the fastest way to find out which parts you actually understand.

Mandi has a view you have to work not to get distracted by.

I picked up drone flying after joining IITGN, and later came back to the same course as a teaching assistant — this time on the other side of it.

Teaching around thirty students to plan a mission, fly it, and bring back data worth processing.

Out with the Earth Science group for drone and core acquisition. Three campaigns between 2023 and 2024.

Fieldwork teaches you things about your own data that no amount of processing will.