Investigating face pareidolia using DeepGaze: Bridging human and artificial perception [In Preparation]
Gupta, P., & Dobs, K. (2024). "Investigating face pareidolia using deepgaze: Bridging human and artificial perception [In Preparation]."
CV
Education
- Ph.D. in Visual Cognition & Computational Neuroscience, Justus-Liebig University (JLU), Giessen, Germany, 2020 - 2024 (expected)
- Thesis: “From Faces to Objects: Exploring the boundaries of artificial and human perception.”
- M.Sc. in Simulation Science, Rheinisch-Westfälische Technische Hochschule (RWTH), Aachen, Germany, 2017 – 2020
- Focus: Computational modeling with a focus on data-driven applications.
- B.Tech. in Mechanical Engineering, National Institute of Technology (NIT), Kurukshetra, India, 2012–2016
- Focus: Design, simulation, production, and operations.
Work experience
- December 2020 - Spring 2024: Ph.D. Researcher
- Justus-Liebig University (JLU), Giessen, Germany
- Duties include: Conducting research on the computational aspects of visual cognition and developing neural network models to investigate human and artificial perception of faces and objects.
- Supervisor: Dr.rer.nat. Katharina Dobs (katharina.dobs@gmail.com)
- April 2018 - September 2020: Hilfswissenschaftler (Part-Time Software Developer)
- Fraunhofer Institute of Production Technology (IPT), Aachen, Germany
- Duties included: Developing predictive tool wear models using deep learning and improving manufacturing processes through data analysis and visualization.
- Supervisor: Mr. M.Sc. Carsten Holst (carsten.holst91@gmail.com)
- August 2016 - August 2017: Technical Planner (Assistant Manager)
- Tata Motors Limited (TML), Rudrapur, India
- Duties included: Implementing machine learning techniques for process automation and enhancing manufacturing efficiency.
- Supervisor: Mr. Sudhanshu Mishra (linkedin id: sudhanshu-mishra-916843180)
Skills
- Programming Languages
- Python, Matlab, C/C++, Bash Scripting, Java (design-level familiarity)
- Data Technologies
- SQL, MySQL/PostgreSQL/LiteSQL/MongoDB, Docker, Hadoop, MapReduce
- AI Frameworks
- PyTorch, TensorFlow, Keras, Pandas, NumPy, SciPy, OpenCV
- Version Control
- GitHub, GIT Source Version Control, GitLab
Publications
Human-like face pareidolia emerges in deep neural networks optimized for face and object recognition [Under Review]
Gupta, P., & Dobs, K. (2023). "Human-like face pareidolia emerges in deep neural networks optimized for face and object recognition [Under Review]."
CNNs reveal the computational implausibility of the expertise hypothesis
Kanwisher, N., Gupta, P., & Dobs, K. (2023). "CNNs reveal the computational implausibility of the expertise hypothesis." Iscience, 26(2).
Deep learning and rule-based image processing pipeline for automated metal cutting tool wear detection and measurement
Holst, C., Yavuz, T. B., Gupta, P., Ganser, P., & Bergs, T. (2022). "Deep learning and rule-based image processing pipeline for automated metal cutting tool wear detection and measurement." IFAC-PapersOnLine, 55(2), 534–539.
Digital image processing with deep learning for automated cutting tool wear detection
Bergs, T., Holst, C., Gupta, P., & Augspurger, T. (2020). "Digital image processing with deep learning for automated cutting tool wear detection." JProcedia Manufacturing, 48, 947–958.
Talks
Investigating face pareidolia using DeepGaze: Bridging human and artificial perception
Poster at CMBB Day 2023, Giessen, Germany
CNNs reveal the computational implausibility of the expertise hypothesis
Poster at ECVP 2022, Nijmegen, Netherlands
Human-like face pareidolia emerges in deep neural networks optimized for face and object recognition
Talk at SFB Retreat 2022, Schloss Rauischholzhausen, Germany
CNNs reveal the computational implausibility of the expertise hypothesis
Poster at LIN 2022 Workshop, Schloss Rauischholzhausen, Germany
A computational explanation for the unreasonable human ability to detect faces in things
Talk at ECVP 2021, Online (Covid Lockdown)
Teaching
Service and leadership
- Currently signed in to multiple slack, discord and mattermost teams