Ntumba Elie, Nsampi

I am a Ph.D. student at the Max Planck Institute For Informatics and Saarland University, where i am fortunate to be supervised by Dr. Thomas Leimkühler. Prior to joining MPI-INF i obtained a masters in Computer Science from Northwestern Polytechnical University where i was supervised by Professor Qing Wang.

Email  /  CV  /  Linkedin

profile photo
Research

I am broadly interested in Computer Vision, Computer Graphics and, Machine Learning. I am particularly interested in Problems at their Intersection.

Neural Field Convolutions by Repeated Differentiation
Ntumba Elie, Nsampi, Adarsh Djeacoumar, Hans-Peter Seidel, Tobias Ritschel, Thomas Leimkühler,
ACM TOG (SIGGRAPH Asia), 2023
Project, Paper

We introduce an algorithm to perform efficient continuous convolution of neural fields.

SIDNet: Learning Shading-aware Illumination Descriptor for Image Harmonization
Zhongyun Hu, Ntumba Elie Nsampi,Xue Wang, Qing Wang
IEEE TETCI, 2023
Project, Paper

A new framework, with the first image harmonization dataset that has shading variations.

Physically Inspired Neural Rendering for any-to-any Relighting
Zhongyun Hu, Ntumba Elie, Nsampi, Xue Wang, Qing Wang
IEEE Transaction on Image Processing, 2022
Paper

We decompose the any-to-any relighting problem, into three sub-problems and propose three networks to solve each one independently.

Neural Shading Field for Image Harmonization
Zhongyun Hu, Ntumba Elie, Nsampi,Xue Wang, Qing Wang
Arxiv, 2021
Paper / Code (Coming soon)

Learning Exposure Correction Via Consistency Modeling
Ntumba Elie, Nsampi, Zhongyun Hu, Qing Wang
BMVC, 2021
Paper / Code

We constrain a deep network to learn an exposure-invariant representation, such that images of different exposure degradation level result in the same representation.

Depth Guided Image Relighting Challenge
Ntumba Elie, Nsampi, Zhongyun Hu, Qing Wang
Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), 2021
Paper

We propose a shadow guidance network, which when plugged into an any-to-any relighting pipeline improves the quality of generated shadows.


Thanks to Jon Barron for the website template.
Last updated September 2023.