Khurram Azeem Hashmi

Researcher at DFKI and Ph.D. Candidate at RPTU Kaiserslautern

Google Scholar Linkedin CV Twitter

About Me

I am a researcher at the German Research Center for Artificial Intelligence (DFKI) and a PhD Candidate at RPTU Kaiserlautern-Landau, where I work closely with Prof. Dr. Didier Stricker on developing new approaches for visual recognition. I obtained my bachelor's degree in Computer Science from University of Computer and Emerging Sciences, Pakistan, in 2016, and the M.S. degree and my master's degree in Artificial Intelligence from the Technical University of Kaiserslautern.

My research focuses on developing algorithms and models that enable machines to "see" and understand the visual world. I'm particularly interested in applying deep learning techniques to solve problems in computer vision, such as imporving instance-level representations in Videos and other challenging ennvironments. In my research, I explore ways to make these models more efficient, robust, and interpretable. In my free time, I enjoy tinkering with new machine learning models, reading novel research methods, and exploring the great outdoors. I'm always on the lookout for new challenges and opportunities to learn, so feel free to get in touch with me.

News

[May 2024] Code for FeatEnHancer has been released and can be accessed through this Link.

[March 2024] Sparse Semi-Detr: Sparse Learnable Queries for Semi-Supervised Object Detection is accepted at CVPR'24. Many congratulations to co-authors. More details will follow.

[July 2023] One paper on Learning Instance Representation for downstream vision tasks under Low-light is accepted at ICCV'23.

[June 2023] A comprehensive survey on Object Detection with Transformers is released on arXiv. Besides the preprint, a
unified reposoitry is published to track the advancements on transformer-based object detection methods.

[February 2023] I feel honoured to announce that I am nominated for an AI-Newcomer 2023 award by the German Society of Computer Science. Please Vote now by clicking Here.

[October 2022] One paper on Vido Object detection is accepted at WACV'23.

[August 2022] One paper on Spatio-temporal learnable proposals for Vido Object detection is accepted at BMVC'22.

Selected Publications [Google Scholar]

	FeatEnHancer: Enhancing Hierarchical Features for Object Detection and Beyond Under Low-Light Vision K.A. Hashmi, G. Kallempudi, D.Stricker, M.Z. Afzal ICCV 2023 [PDF] [Webpage] [Code]
	Object Detection with Transformers: A Review T. Shehzadi K.A. Hashmi, D.Stricker, M.Z. Afzal arXiv 2023 [PDF] [Code]
	BoxMask: Revisiting Bounding Box Supervision for Video Object Detection K.A. Hashmi, A.Pagani. D.Stricker, M.Z. Afzal WACV 2023 [PDF] [Code]
	Spatio-Temporal Learnable Proposals for End-to-End Video Object Detection K.A. Hashmi, D.Stricker, M.Z. Afzal BMVC 2022 [PDF] [Video]
	Attention-Guided Disentangled Feature Aggregation for Video Object Detection S. Muralidhara, K.A. Hashmi, A.Pagani, D.Stricker, M.Z. Afzal Sensors 2022 [PDF]
	Exploiting Concepts of Instance Segmentation to Boost Detection in Challenging Environments K.A. Hashmi, A.Pagani, M. Liwicki, D.Stricker, M.Z. Afzal Sensors 2022 [PDF]
	Guided Table Structure Recognition Through Anchor Optimization K.A. Hashmi, D.Stricker, M. Liwicki, M.Z. Afzal IEEE ACCESS 2022 [PDF]

Services and Reviews

Reviewer of Conferences:

ECCV2024, CVPR2024, WACV2023, ECCV2022, BMVC2022

Reviewer of Journals:

Honors & Awards

Received merit based full scholarship for complete Bachelors
Nominated for the best AI Newcomer Award by German Assoication of Computer Science
Nominated for the AI Person of the Year Award 2024 by data:unplugged and t3n Magazin

Reviewer of Conferences:

Reviewer of Journals:

Received merit based full scholarship for complete Bachelors

Nominated for the best AI Newcomer Award by German Assoication of Computer Science

Nominated for the AI Person of the Year Award 2024 by data:unplugged and t3n Magazin