Lokender Tiwari

Department of Computer Science and Engineering
IIIT-Delhi

Okhla Industrial Estate,Phase III, New Delhi, India - 110020

Email: lokendert@iiitd.ac.in
[Google Scholar] [LinkedIn] [Github]

About me

I am a Ph.D. Candidate at , IIIT-Delhi. I work in the domain of Computer Vision and Deep Learning. I am advised by Dr. Saket Anand. My Ph.D. is supported by Visvesvarya Ph.D. Fellowship. During summer 2019, I was Ph.D. Research Intern at Media Analytics Department of NEC Laboratories America, Inc. I was working with Prof. Manmohan Chandraker, Dr. Quoc-Huy Tran, and Dr. Pan Ji.

Research Interest

I am interested in developing robust methods for low-level geometric scene understanding (e.g SfM/SLAM, multiple-geometric model fitting, image-based 3D modeling) and high-level semantic understanding (e.g image classification, object detection, segmentation). In general, for robust scene understanding, I am interested in developing methods that can handle noise, outliers, and adversarial perturbations. Recently, I focus on the following research topics:

News

Work Experiance

Awards

Patent Applications

 

Pseudo-RGB-D: Exploiting Learned Depth for Robust SfM

Quoc-Huy Tran, Lokender Tiwari, Pan Ji, Manmohan Chandraker
NEC Laboratories America, Inc.
(Application #US/16/987,705, filed on Aug 2020)

Publications

 

Pseudo RGB-D for Self-Improving Monocular SLAM and Depth Prediction

Lokender Tiwari, Pan Ji, Quoc-Huy Tran, Bingbing Zhuang, Saket Anand, Manmohan Chandraker
European Conference on Computer Vision (ECCV), 2020 (Accepted).
[Preprint] [Project Page] [BibTex] [Demo 1 ] [Demo 2 ] [Demo 3]

 

Dissecting Deep Networks into an Ensemble of Generative Classifiers for Robust Predictions

Lokender Tiwari, Anish Madan, Saket Anand, Subhashish Banerjee
arXiv, 2020.
[Preprint] [Project Page] [Code] [BibTex]

 

DGSAC: Density Guided Sampling and Consensus (Journal version: with new fully automated multi model hypothesis sampling algorithm and Quadratic program based model selection.)

Lokender Tiwari, Saket Anand
arXiv, 2020.
[Preprint] [Code coming soon] [BibTex]

 

DGSAC: Density Guided SAmpling and Consensus

Lokender Tiwari, Saket Anand
IEEE Winter Conference on Computer Vision (WACV), 2018.
[PDF Author Version] [Code] [BibTex]

 

Robust Multi-Model Fitting using Density and Preference Analysis

Lokender Tiwari, Saket Anand, Sushil Mittal
Asian Conference on Computer Vision (ACCV), 2016.
[PDF Author Version] [Code] [BibTex]

 

Fast Hypothesis Filtering for Multi-Structure Geometric Model Fitting

Lokender Tiwari, Saket Anand
IEEE International Conference on Image Processing (ICIP), 2016.
[PDF Author Version] [BibTex]

 

Robust Camera Captured Image Mosaicking for Document Digitization and OCR Processing

Lokender Tiwari, Bhupendra Kumar, Tushar Patnaik
IEEE International Conference on Information Technology (ICIT), 2014.
[DOI] [BibTex]

PhD Forum/Symposium/Workshops

 

Robust Statistical Methods for Learning and Vision

Lokender Tiwari
Mentor: Prof. Terrance Boult
PhD Forum, IEEE Winter Conference on Computer Vision (WACV), 2014.

 

Robust Estimation in Geometric Computer Vision

Lokender Tiwari
Doctoral Symposium (Best Doctoral Consortium Award (3rd))
Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP), 2020.

 

Geometric Model Fiting Game

Lokender Tiwari
Symposium on Optimization at IIT-Kanpur, 2020.

Professional Services

Volunteer:

  • International Conference on Learning Representations (ICLR) 2020
  • International Conference on Machine Learning (ICML) 2020

Reviewer:

  • Elsevier Pattern Recognition
  • IEEE Transactions on Industrial Electronics
  • IEEE Access
  • IEEE Winter Conference on Computer Vision (WACV)

Collaborators