Biography
I am a third-year Ph.D. student in the
University of Munich on Deep Learning, advised by
Prof. Volker Tresp.
My research interests include computer vision and deep learning. Especially, I am interested in the following topics:
- Scene Graph and Scene Understandings
- Visual Question Answering using Scene Graph Reasoning
- Image and Object Understanding with Cognitive Models (e.g. Transformers)
- Semi/Self-supervised Representation Learning for Outlier Detection
News
-
[Jul. 2022] Our paper Relationformer, a single stage unified image to graph generation got accepted in ECCV 2022.
-
[Dec. 2021] An analytical paper of us on contrastive learning and its connection towards cluster for OOD detection has been accepted in AAAI-SafeAI.
-
[Oct. 2021] Our paper OODformer presented a new transformer based approach for OOD detection and got accepted in BMVC 2021.
-
[May. 2021] One paper have been accepted to ISWC 2021.
-
[June. 2020] Two paper accepted in ICML workshop.
Education
Ph.D. Candidate:
Master Degree:
Bachelor Degree:
Work Experience
University of Munich:
- 2018.10 - present: Doctoral Researcher
- University of Munich (LMU), Munich, Germany
- Topic: Scene Graph, Exploration of Context in an Image or Video
Fortiss:
- 2017.01 - 2018.04: Research Intern
- Fortiss GMBH, a Research Institute from TUM, Munich, Germany
- Topic: Comprehensive Fore-sight of Traffic Situation through Object Detection and Tracking
J.P Morgan Chase/ Cognizant:
- 2011.12 - 2015.09: Software Developer
- J.P. Morgan Chase and Cognizant, Kolkata, India
- Topic: an one stop OFAC payment processing and workflow investigation.
Publications
---Scene Graph Generation---
-
Relationformer: A Unified Framework for Image-to-Graph Generation
Suprosanna Shit*,Rajat Koner*,Bastian Wittmann ..., European Conference on Computer Vision(ECCV), 2022; arixv [PDF] [CODE]
-
Relation Transformer Network
Rajat Koner, Suprosanna Shit, Volker Tresp; arixv [PDF] [CODE]
-
Scenes and Surroundings: Scene Graph Generation using Relation Transformer
Rajat Koner, Poulami Sinhamahapatra, Volker Tresp; International Conference on Machine Learning (ICML) workshop, 2020 [PDF]
-
Improving Visual Relation Detection using Depth Maps
Sahand Sharifzadeh, Sina Moayed Baharlou, Max Berrendorf, Rajat Koner, Volker Tresp; International Conference on Pattern Recognition (ICPR), 2020 [PDF]
---Visual Question Answering---
-
Graphhopper: Multi-Hop Scene Graph Reasoning for Visual Question Answering
Rajat Koner, Hang Li, Marcel Hildebrandt, Deepan Das, Volker Tresp and Stephan Günnemann; International Semantic Web Conference(ISWC), 2021 [PDF] [CODE]
-
Scene Graph Reasoning for Visual Question Answering
Marcel Hildebrandt, Hang Li, Rajat Koner, Volker Tresp and Stephan Günnemann; International Conference on Machine Learning (ICML) workshop, 2020 [PDF]
---Out-Of-Distribution Detection(OOD) using Representation Learning---
-
OODformer: Out-Of-Distribution Detection Transformer
Rajat Koner, Poulami Sinhamahapatra, Karsten Roscher, Stephan Günnemann and Volker Tresp; arixv, 2021 [PDF] [CODE]
-
Is it all a cluster game? - Exploring Out-of-Distribution Detection using Cluster-based Distances
Poulami Sinhamahapatra, Rajat Koner, Karsten Roscher,Volker Tresp and Stephan Günnemann and ; arixv, 2021. [PDF]
---Others---
-
Random finite set based Bayesian filtering with OpenCL in a heterogeneous platform
Biao Hu, Uzair Sharif, Rajat Koner, Guang Chen, Kai Huang, Feihu Zhang, Walter Stechele, Alois Knoll; Sensors, MDPI, 2018 [PDF]
Other
-
Conference Reviewer:
- Neural Information Processing Systems (NeurIPS) 2021
- European Conference on Computer Vision (ECCV) 2020
-
Journal Reviewer:
- IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI)
-
Teaching:
- Machine Learning (2021)
- Introduction to Deep Learning (2020)