Jogendra Nath Kundu

Email: jogendranathkundu@gmail.com / Google scholar / Linkedin

I am a Research Scientist at Reality Labs Research, Meta, Redmond, WA, USA. I completed my PhD at Video Analytics Lab, CDS, Indian Institute of Science, under the supervision of Prof. R. Venkatesh Babu. My research primarily focuses on developing self-supervised domain adaptation solutions with a particular interest for dense prediction tasks such as depth estimation, semantic segmentation, and 3D human pose estimation.

Research

Generalization in machine learning refers to the abstract feature of intelligence, which enables an algorithm/model to be effective across a range of inputs and applications. Dependency on high-quality annotations substantially limits the generalizability of supervised deep learning models. Structured prediction-based vision tasks such as scene segmentation or 3D human pose estimation suffer the most. My research aims to overcome these limitations. The key directions include; 1) curating useful synthetic domain data, 2) leveraging metadata relationships for self-supervised learning, and 3) unsupervised domain adaptation for continuously changing environments.

News

Selected Publications

(see Google Scholar for the full list of papers) 

Concurrent Subsidiary Supervision for Unsupervised Source-Free Domain Adaptation

Jogendra Nath Kundu, Suvaansh Bhambri, Akshay Kulkarni, Hiran Sarkar, Varun Jampani, R. Venkatesh Babu. In ECCV 2022.

Uncertainty-Aware Adaptation for Self-Supervised 3D Human Pose Estimation

Jogendra Nath Kundu, Siddharth Seth, Pradyumna YM, Varun Jampani, R. Venkatesh Babu, Anirban Chakraborty. In CVPR 2022.

Non-Local Latent Relation Distillation for Self-Adaptive 3D Human Pose Estimation

Jogendra Nath Kundu, Siddharth Seth, Anirudh Jamkhandi, Pradyumna YM, Varun Jampani, R. Venkatesh Babu, Anirban Chakraborty. In NeurIPS 2021.

Aligning Silhouette Topology for Self-Adaptive 3D Human Pose Recovery

Mugalodi Rakesh, Jogendra Nath Kundu, Varun Jampani, R. Venkatesh Babu. In NeurIPS 2021.

Generalize then Adapt: Source-Free Domain Adaptive Semantic Segmentation

Jogendra Nath Kundu, Akshay Kulkarni, Amit Singh, Varun Jampani, R. Venkatesh Babu. In ICCV 2021.

Appearance Consensus Driven Self-Supervised Human Mesh Recovery

Jogendra Nath Kundu, Mugalodi Rakesh, Varun Jampani, Rahul MV, R. Venkatesh Babu. In ECCV 2020 (Oral presentation, top 2%).

Class-Incremental Domain Adaptation 

Jogendra Nath Kundu, Rahul Mysore Venkatesh, Naveen Venkat, Ambareesh Revanur, R. Venkatesh Babu. In ECCV 2020.

Self-Supervised 3D Human Pose Estimation via Part Guided Novel Image Synthesis 

Jogendra Nath Kundu, Siddharth Seth , Varun Jampani, Mugalodi Rakesh , R. Venkatesh Babu, Anirban Chakraborty. In CVPR 2020 (Oral presentation, top 5.7%).

Towards Inheritable Models for Open-Set Domain Adaptation 

Jogendra Nath Kundu, Naveen Venkat, Ambareesh Revanur, Rahul M V, R. Venkatesh Babu. In CVPR 2020 (Oral presentation, top 5.7%).

Universal Source-Free Domain Adaptation 

Jogendra Nath Kundu, Naveen Venkat, Rahul M V, R. Venkatesh Babu. In CVPR 2020.

Kinematic-Structure-Preserved Representation for Unsupervised 3D Human Pose Estimation 

Jogendra Nath Kundu, Siddharth Seth , Rahul M V, Mugalodi Rakesh , R. Venkatesh Babu, Anirban Chakraborty. In AAAI 2020 (Oral presentation, top 5.8%).

UM-Adapt: Unsupervised Multi-task Adaptation using Adversarial Cross-task Distillation

Jogendra Nath Kundu, Nishank Lakkakula, R. Venkatesh Babu. In ICCV 2019 (Oral presentation, top 4.3%).

GAN-Tree: An Incrementally Learned Hierarchical Generative Framework for Multi-modal Data

Jogendra Nath Kundu, Maharshi Gor, Dakshit Agrawal, R. Venkatesh Babu. In ICCV 2019.

BiHMP-GAN: Bidirectional 3D Human Motion Prediction GAN

Jogendra Nath Kundu, Maharshi Gor, R. Venkatesh Babu. In AAAI 2019 (Spotlight presentation).

iSPA-Net : Iterative Semantic Pose Alignment Network

Jogendra Nath Kundu, Aditya Ganeshan, Rahul M V, Aditya Prakash, R. Venkatesh Babu. In ACM Multimedia (ACMMM) 2018 (Spotlight presentation).

AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimation

Jogendra Nath Kundu,  Phani Krishna Uppala, Anuj Pahuja, R. Venkatesh Babu. In Computer Vision and Pattern Recognition (CVPR) 2018 (Spotlight presentation).

Enabling my robot to play Pictionary: Recurrent Neural Networks for sketch recognition

Ravi Kiran Sarvadevabhatla, Jogendra Nath Kundu, R. Venkatesh Babu. In  ACM Multimedia (ACMMM) 2016.

Study of feature combination using HMM and SVM for multilingual Odiya speech emotion recognition

Monorama Swain, Subhasmita Sahoo, Aurobinda Routray, Prithviraj Kabisatpathy, Jogendra Nath Kundu. In  International Journal of Speech Technology (IJST), Sep 2015.