Skip to main content Link Search Menu Expand Document (external link)

Tejaswi Kasarla

I am interested in problems involving computer vision and machine learning with real world applications. I am a third-year PhD at VIS lab, University of Amsterdam and ELLIS Unit Amsterdam, advised by Pascal Mettes and Rita Cucchiara. During my PhD, I will explore how to leverage prior knowledge and inductive biases in learning visual data, and if that intersts you, read more about my thesis topic on the ELLIS website.

I was one of the organizers of Women in Computer Vision (WiCV) Workshop, co-hosted with CVPR 2021 and CVPR 2022. Currently, I’m serving as a board member for WiCV (since Nov 2022). In my free time, I enjoy playing ukulele and reading books!

Email | Google Scholar | CV | Github | LinkedIn | Twitter

highlights

June 2024Visiting Rita Cucchiara at University of Modena and Reggio Emilia until September 2024.
2023Reviewed papers for ICCV 2023, WiCV Workshop @ CVPR 2023, NeurIPS 2023 and ICLR 2024.
Nov 2022Teaching Assistant for Applied Machine Learning course.
Sep 2022I presented a poster on our NeurIPS 2022 paper the ELLIS Doctoral Symposium 2022 in Alicante, Spain!
Sep 2022Our work on Maximum Separation is accepted to NeurIPS 2022 ! I will be presenting this work in-person at the conference. I will also present this work as a contributed talk at the Women in Machine Learning workshop.
Jun 2022New preprint on arXiv!: Maximum Class Separation as Inductive Bias in One Matrix
Jun 2022We organized the in-person and virtual Women in Computer Vision Workshop at CVPR 2022!
Nov 2021Teaching Assistant for Applied Machine Learning course
Oct 2021I joined PhD at VIS Lab, University of Amsterdam.
Oct 2021We organized the Women in Computer Vision (WiCV) Social at ICCV 2021!
Jun 2021We organized the Women in Computer Vision Workshop at CVPR 2021!
Aug 2020Presented poster on Region based Active Learning for Semantic Segmentation at Summer School of Machine Learning at Skoltech (SMILES)
Aug 2019Reviewed papers for Women in Machine Learning (WiML) workshop, co-located with NeurIPS 2019.
Jun 2019Defended my Master’s thesis!
May 2019Started working as Computer Vision Researcher at Research and Technology Center, Bosch. Excited to work on problems related to Autonomous Driving. | Collaborators: Amit Kale, Yasaswi Bharadwaj, Hiranmai M., Subramanya Bharadwaj
Nov 2018Paper on Region Based Active Learning for Efficient Labelling in Semantic Segmentation accepted to WACV 2019 [link]
Jun 2018 - Oct 2018Interning at Research and Technology Center, Bosch. | Mentors: Guruprasad Hegde, Amit Kale
Apr 2018Presenting our work on Active Learning for Semantic Segmentation at 1st Research Symposium, IIIT Hyderabad

publications

Maximally Separated Active Learning
Tejaswi Kasarla, Abhishek Jha, Faye Tervoort, Rita Cucchiara, Pascal Mettes
ECCV 2024 Beyond Euclidean Workshop

code

Lightweight Uncertainty Quantification with Simplex Semantic Segmentation for Terrain Traversability Judith Dijk, Gertjan J. Burghouts, Kapil D. Katyal, Bryanna Y. Yeh, Craig T. Knuth, Ella Fokkinga, Tejaswi Kasarla, Pascal Mettes
ICRA 2024 Workshop on Resilient Off-road Autonomy

paper

Maximum Class Separation as Inductive Bias in One Matrix
Tejaswi Kasarla, Gertjan J. Burghouts, Max van Spengler, Elise van der Pol, Rita Cucchiara, Pascal Mettes
NeurIPS 2022 (Oral – top 0.08%)

Also a contributed talk at WiML Workshop @ NeurIPS 2022

project pagepaper code poster media

Region-Based Active Learning for Efficient Labelling in Semantic Segmentation
Tejaswi Kasarla, G Nagendar, Guruprasad Hegde, Vineeth N. Balasubramanian, C.V. Jawahar
WACV 2019

paper supplementary poster