Best Student Paper Award
Camera Calibration through Camera Projection Loss @ ICASSP 2022
So, today I received the Best Student Paper Award at ICASSP 2022 for my work on Camera Calibration through Camera Projection Loss. The story behind the work goes as follows:
It was 10th February, 2019 when I first met Dr. Murtaza Taj with an aim of doing some research work and then extending it as my MS Thesis. We discussed different projects and ultimately locked Cross View Image Retrieval as the starting point. Within six months we got it published at ICONIP-2019.
By the time, our work got published, I had met Faisal Bhai and was motivated to publish a paper as first author for which I decided to work on my own idea.
I went to Dr. Murtaza and told him that I want to work on my own idea to which surprisingly he agreed and I chose Cyclist Detection as the problem at hand but was unable to detect small objects.
I again went to Dr. Murtaza asking him to work on yet another problem, to which he again agreed and selected Camera Calibration as the problem motivated by the work of Oneeb.
For Camera Calibration, I generated a new dataset using CARLA and embedded mathematical equations in a CNN which ultimately got accepted at ICASSP 2022 after some rejections at different conferences.
Yesterday, we received an email stating that the ICASSP 2022 best paper and best student paper awards were being finalized and that the winners of these paper awards (best papers, best student papers, best paper award for industry) will be announced in the closing ceremony on 27 May ’22 and we were asked to join it.
Today, we received an email stating that we will receive an award during the paper award ceremony. There were 6 best student paper awards and 2 best paper awards.
I wasn’t expecting the award and believe that it wouldn’t have been possible without the support of Dr. Murtaza Taj and my family, specially my mother. I don’t have any words to express my feelings so that’s it for now. See you later.
The paper can be accessed here.
Some related articles are as follows: