Publication(s)

2020
Weakly-Supervised 3D Human Pose Learning via Multi-view Images in the Wild
Umar Iqbal, Pavlo Molchanov,  Jan Kautz
IEEE Conference on Computer Vision and Pattern Recognition (CVPR'20), Seattle, USA, 2020. 
[PDF]​
​
Self-Supervised Viewpoint Learning from Image Collections
​Siva Karthik Mustikovela, Varun Jampani, Shalini De Mello, Sifei Liu, Umar Iqbal, Carsten Rother, Jan Kautz
IEEE Conference on Computer Vision and Pattern Recognition (CVPR'20), Seattle, USA, 2020. 
[PDF]​[Code]
​
Weakly-Supervised 3D Hand Pose Estimation via Biomechanical Constraints 
Adrian Spurr, Umar Iqbal, Pavlo Molchanov, Otmar Hilliges, Jan Kautz
arXiv preprint 2020 
[PDF]​

2019

Few Shot Adaptive Gaze Estimation
Seonwook Park, Shalini De Mello, Pavlo Molchanov, Umar Iqbal, Otmar Hilliges, Jan Kautz
International Conference on Computer Vision (ICCV'19), Seoul, South Korea, 2019
[PDF]​[Code]

2018

Articulated Human Pose Estimation in Unconstrained Images and Videos
Dissertation, University of Bonn, Germany, 2018.
Summa cum laude, DAGM MVTec Dissertation Award 2019
[PDF]​​[Summary][Slides][Talk]
Hand Pose Estimation via Latent 2.5D Heatmap Regression
U. Iqbal, P. Molchanov, T. Breuel, J. Gall and J. Kautz
European Conference on Computer Vision (ECCV'18), Munich, Germany, 2018 
Invited to the 4th Workshop on Observing and Understanding Hands in Action (HANDS), 2018
Best Poster Award @ HANDS'18
[PDF]​​[Video][Poster​][Plots]
JointFlow: Temporal Flow Fields for Multi Person Pose Tracking
A. Doering, U. Iqbal, J. Gall
British Machine Vision Conference (BMVC'18), Newcastle, UK, 2018
[PDF]​​[Poster]
PoseTrack: A Benchmark for Human Pose Estimation and Tracking
M. Andriluka, U. Iqbal, A. Milan, E. Insafutdinov, L. Pishchulin,  J. Gall and B. Schiele
IEEE Conference on Computer Vision and Pattern Recognition (CVPR'18),  Salt Lake City, USA, 2018
[PDF]​​[Project Page][Video][Data][Code]
A Dual-Source Approach for 3D Human Pose Estimation from Single Images 
U. Iqbal, A. Doering, H. Yasin, B. Krüger, A. Weber, and J. Gall
Computer Vision and Image Understanding (CVIU), 2018
[PDF]​[Project Page][Code]

2017

PoseTrack: Joint Multi-Person Pose Estimation and Tracking
U. Iqbal, A. Milan, and J. Gall
IEEE Conference on Computer Vision and Pattern Recognition (CVPR'17),  Hawaii, USA, July 2017
[PDF][Bibtex][Project Page][Video][Data][Code]
Pose for Action - Action for Pose
U. Iqbal, M. Garbade, and J. Gall
IEEE Conference on Automatic Face and Gesture Recognition (FG'17), Washington-DC, USA, 2017
​[PDF][Bibtex][Project Page][Code]​

2016

Multi-Person Pose Estimation with Local Joint-to-Person Associations
U. Iqbal, and J. Gall
Crowd Understanding Workshop (CUW). In conjunction with ECCV'16, Amsterdam, 2016 
[PDF][Bibtex][Project Page][Code]
A Dual-Source Approach for 3D Pose Estimation from a Single Image
H. Yasin*, U. Iqbal*, B. Krüger, A. Weber, and J. Gall (*equal contribution)
IEEE Conference on Computer Vision and Pattern Recognition (CVPR'16),  Las Vegas, USA, June 2016.
[PDF] [Bibtex]​[Project Page]​[Code]

Older publications

Who is the Hero? - Semi-Supervised Person Re-Identification in Videos 
​U. Iqbal, I. D. D. Curcio, and M. Gabbouj,
Int. Conference on Computer Vision Theory and Applications (VISAPP'14), Lisbon, Portugal, 2014.  
[PDF][Bibtex]
Image Based Vehicle Type Identification
U. Iqbal, S.W. Zamir, M.H. Shahid, K. Parwaiz, M. Yasin and M.S. Sarfraz
IEEE International Conference on Information and Emerging Technologies (ICIET'10, Oral),  
​Pakistan, 2010. 
​​[PDF][Bibtex][Data]

Previous Projects

Important person detection from multiple videos (Master's Thesis)

The goal of this work is to develop a unified framework for unsupervised person re-identification with application to important person detection from multiple videos. It focuses on topics such as
  • Face detection
  • Particle filters based object tracking
  • Face/clothing recognition
  • Detection of high level attributes (i.e. Race, Gender, Eyewears, Headwears, Hair colors, Age, etc.). An example can be seen here.
  • Sensor based temporal segmentation of videos
  • Supervised classification.
  • Hierarchical Agglomerative Clustering (HAC) with constraints (i.e. cannot-link constraints)
Tools used in this work: C++, OpenCV, Python, Intel's Threading Building Blocks (TBB), GNU Scientific Library (GSL)
Link: PDF
Thesis Grade: 5 out of 5

Vehicle make and model recognition (Bachelor's Thesis)

I worked on this topic as my undergraduate thesis and also followed the same work during my tenure at COMVis. The results of the work were published in ICIET 2010 and the paper can be found here.
Vehicle's database used in this work can be downloaded from this link.
Demo: 

Preceding Vehicle Detection and Distance Estimation for Driver Assistance System

Technical Report on the project can be found here. (Last updated: 31-12-2010)
Demo: