Research Interests

Computer Vision | Computer Graphics | Deep Learning | Graph Representation Learning

It's all about learning to get (surface) Normal!

My primary research interest is in extending the boundaries of photometric methods - Photometric Stereo, Shape from Polarization, and Photo-polarimetric Stereo to infer the physical 3D world through (neural) inverse rendering approaches. I am also interested in exploring creative art forms, human pose transfer, and graph representation learning. 

Publications

SS-SfP: Neural Inverse Rendering for Self-Supervised Shape from (Mixed) Polarization (Oral)

Ashish Tiwari and Shanmuganathan Raman 

To appear in the Pacific Graphics 2023, Daejeon, South Korea, October 10-13, 2023.

paper | supp

Hand Shadow Art: A Differentiable Rendering Perspective (Best Poster Award)

Aalok Gangopadhyay, Prajwal Singh, Ashish Tiwari and Shanmuganathan Raman 

To appear in the Pacific Graphics 2023, Daejeon, South Korea, October 10-13, 2023.

paper

TreeGCN-ED: A Tree-Structured Graph-Based Autoencoder

Framework For Point Cloud Processing

Prajwal Singh, Ashish Tiwari, Kaustubh Sadekar, and Shanmuganathan Raman 

To appear in the Pacific Graphics 2023, Daejeon, South Korea, October 10-13, 2023.

project page

Revisiting Heterophily in Graph Convolutional Networks By Learning Representations Across Topological and Feature Spaces

Ashish Tiwari, Sresth Tosniwal, and Shanmuganathan Raman 

project page | paper

Deep Appearance Consistent Human Pose Transfer

Ashish Tiwari, Zeeshan Khan, Aditya Vora, Manjuprakash Rama Rao, and Shanmuganathan Raman 

International Conference on Pattern Recognition (ICPR) 2022, Montreal, Quebec, Canada, August 21-25, 2022.