Akshay L Chandra bio photo

Akshay L Chandra

M.Sc. Student at
University of Freiburg
Freiburg im Breisgau, DE

Google Scholar Github Medium LinkedIn

Welcome

Hi! I am an M.Sc. Computer Science student at the University of Freiburg, specializing in Artificial Intelligence. Earlier, I was a research assistant at Prof. Vineeth N Balasubramanian’s research group Lab1055 at IIT Hyderabad for nearly 3 years, mainly working on deep active learning for image data. My interests are in the intersection of computer vision and learning with limited labeled data. During my tenure at IIT Hyderabad, I also frequently collaborated with Prof. Wei Guo at the Field Phenomics Lab (University of Tokyo) while working on plant phenotyping solutions.

Before joining Dr. Vineeth’s group in December 2018, I spent 16 months working on machine learning and data science projects at GGK Technologies, a service-based company in Hyderabad. At GGK Technologies, I spent most of my time learning, implementing, and optimizing various machine learning/deep learning pipelines for clients in US health care and retail domains.

News

[Jan 2022] Our paper, How Useful is Image-Based Active Learning for Plant Organ Segmentation? is accepted to be included on Plant Phenomics, a Science Partner Journal.
[Aug 2021] After nearly 3 years at IIT Hyderabad, I walk away with many friends that will likely stay in life for a long time and an extraordinary mentor who taught me everything I know about computer science research.
[Jul 2021] Served as a subreviewer for Neural Information Processing Systems conference (NeurIPS'21).
[Jul 2021] Our final paper with results, On Initial Pools for Deep Active Learning is accepted to be included in the Proceedings of Machine Learning Research (PMLR), a sister publication to the Journal of Machine Learning Research (JMLR). This was made possible by the NeurIPS 2020 Preregistration Workshop which encourages an alternative publication model for machine learning research.
[Jun 2021] I married the love of my life. 😊❤️
[Nov 2020] Our proposal, On Initial Pools for Deep Active Learning is accepted at Preregistration Workshop at Neural Information Processing Systems conference (NeurIPS'20).
[Aug 2020] Our toolkit, EasyRFP, An Easy-to-Use Edge Computing Toolkit for Real-Time Field Phenotyping is accepted as an extended abstract at Computer Vision Problems in Plant Phenotyping Workshop (CVPPPW-ECCV) and as an academic demostration at European Conference on Computer Vision, 2020 (ECCV'20).
[Jul 2020] Served as a subreviewer for Neural Information Processing Systems conference (NeurIPS'20) and as a reviewer for Computer Vision Problems in Plant Phenotyping Workshop (CVPPPW-ECCV).
[May 2020] Served as a subreviewer for European Conference on Computer Vision, 2020 (ECCV'20).
[Feb. 2020] Our paper, Active Learning with Point Supervision for Cost-Effective Panicle Detection in Cereal Crops, has been accepted to BMC Plant Methods journal. [Impact Factor - 4.993]
[Jul 2019] Our paper, An Adaptive Supervision for Active Learning in Object Detection, has been accepted to British Machine Vision Conference (BMVC'19) as a poster presentation
[Apr 2019] Released an easy-to-use python library, Image Bounding Box Slicer that slices images and their bounding box annotations into tiles, both into specific sizes and into any arbitrary number of equal parts.
[Dec 2018] Joined Dr. Vineeth N Balasubramanian's research group at IIT-Hyderabad.