Partha Das
PhD candidate in Computer Vision Lab, Informatics Institute, University of Amsterdam
Partha’s piece of the Internet
I hold a PhD from the Computer Vision Lab (Thesis available here), under the Informatics Institute of the University of Amsterdam. I was supervised by Prof. dr. Theo Gevers and dr. Sezer Karaoglu. My research interests are in physics based image understanding, using deep learning & traditional approaches. Specifically, tasks like image formation, light estimation and colour correction, etc. I am interested in levying these approaches to build efficient & stable AR/VR/Mixed Reality systems.
News
2023: Exploratory work on Transformers and physics based learning accepted at ICCV New Ideas in Vision Transformers, Paris!
2023: I defended my PhD thesis (nominated for cum-laude) successfully at the University of Amsterdam!
2022: I’m at ECCV2022. Come say hi! Code for SIGNet is now available!
2022: Two papers are accepted in CVIU and ECCV Workshop CV4Metaverse, respectively!
2022: Accepted at International Computer Vision Summer School.
2022: Code for PIENet is now available.
2021: Three papers are accepted in IJCV, WACV and ICCV Workshop PBDL, respectively!
2021: Winner of Fujitsu AI Hackathon.
2020: Paper accepted at BMVC!
2018: Paper accepted at ECCV!
2017: Winner at Global AI Hackathon, Amsterdam Chapter.
2016: Finalists for the Watson Innovation competition.
Publications
-
IDTransformer: Transformer for Intrinsic Image Decomposition, Accepted in International Conference on Computer Vision (ICCV) Workshop on New Ideas in Vision Transformers (NiVT), 2023. Publication Link
-
SIGNet: Intrinsic Image Decomposition by a Semantic and Invariant Gradient Driven Network for Indoor Scenes, Accepted in European Conference of Computer Vision (ECCV) CV4Metaverse Workshop, 2022. Publication Link
-
PIENet: Photometric Invariant Edge Guided Network for Intrinsic Image Decomposition, Accepted at the Conference on Computer Vision and Pattern Recognition (CVPR), 2022. Publication Link
-
Intrinsic Image Decomposition using Physics-based Cues and CNNs, Accepted in Computer Vision and Image Understanding (CVIU), 2022 Publication Link
-
ShadingNet: Image Intrinsics by Fine-Grained Shading Decomposition, Accepted in International Journal of Computer Vision (IJCV), 2021. (Extension of Master’s Thesis.) Publication Link
-
Generative Models for Multi-Illumination Color Constancy, Accepted in 3rd IEEE International Conference on Computer Vision (ICCV) Workshop on Physics Based Vision meets Deep Learning (PBDL), 2021. Publication Link
-
Eden: Multimodal synthetic dataset of enclosed garden scenes, Accepted in Winter Conference on Applications of Computer Vision (WACV), 2021. Publication Link
-
Novel View Synthesis from Single Images via Point Cloud Transformation. Accepted in British Machine Vision Conference (BMVC), 2020. Publication Link
-
Color Constancy by GANs: An Experimental Survey. Preprint available on arXiv. Publication Link
-
Joint Learning of Intrinsic Images and Semantic Segmentation. Accepted in European Conference on Computer Vision (ECCV) 2018. Publication Link
[expand]
-
A Real-Time Face Motion Based Approach towards Modelling Socially Assistive Wireless Robot Control with Voice Recognition, published in the International Journal of Advanced Computer Science and Applications (IJACSA), 6(10), 2015. Publication Link
-
A Hybrid color plane approach towards Color based Object detection and Modelling of a Real-time Gesture based Intelligent Virtual Aid using Artificial Neural Network, Accepted & Presented at First International Conference on Computing, Communication, Control and Automation, 2015. Publication Link
-
A Novel Approach Towards Detecting Faces and Gender Using Skin Segmentation and Template Matching, Accepted & Presented at Second International Conference on Signal Processing and Integrated Networks, SPIN-2015, Noida, India. Publication Link
-
Estimation of Arm Joint Angles from Surface Electromyography signals using Artificial Neural Networks, published in International Organisation of Scientific Research (IOSR) Journals, Vol. 15, Issue 6, PP. 38-44, Nov.-Dec., 2013. Publication Link
-
A Novel approach for Graphical User Interface development and real time Object and Face Tracking using Image Processing and Computer Vision Techniques implemented in MATLAB, published in International Organisation of Scientific Research (IOSR) Journals, Vol. 15, Issue 5, PP. 61-68, Nov.-Dec., 2013. Publication Link
[/expand]
Academic Experiences
2020 - 2024: Reviewer - ICCV, CVPR, ECCV, TIP, TPAMI, CVIU, Computer & Graphics.
2019 - 2021: Teaching Assistant - Computer Vision 1, University of Amsterdam.
2019 - 2022: Teaching Assistant - Computer Vision 2, University of Amsterdam.
2018 - 2023: MSc. Thesis Supervision.
2018 - 2019: Master’s Project AI Supervision.
2017: BSc. Thesis Supervision.
Thesis Supervision
2023: David Simon Tetruashvili
2022: Friso Verweij, Angelo de Groot, Hannah Min, Dario Shehni Abbaszadeh, Roy van Bragt
2021: Shane Koppers, Gawan Dekker
2020: Tycho Koster, Jeroen Terstall
2019: Masoumeh Bakhtiariziabari, R.P.W. van den Hoogen, Lukas Jelinek, Gururaj Rao, Melissa Tjhia, Yachu Yang
2018: Alexander Albrecht
Find me on the Internet!
Thanks to Wei Wang for the template!