Robin Hesse

PhD Candidate · Visual Inference Lab · TU Darmstadt

I am a PhD candidate in the Visual Inference Lab at TU Darmstadt. My research focuses on narrowing the gap between human and machine intelligence, especially in the context of visual perception. To this end, I study methods of explainable artificial intelligence.


News

· Our paper "Benchmarking the Attribution Quality of Vision Models" has been accepted at NeurIPS 2024 Datasets and Benchmarks Track.
September, 2024
· Honored to be recognized as an Outstanding Reviewer for ECCV 2024.
September, 2024
· I will be attending the International Computer Vision Summer School 2024 in Sicily.
July, 2024
· Our workshop Explainable Computer Vision: Where are We and Where are We Going? has been accepted at ECCV 2024 in Milano. I am looking forward to meeting many people who are interested in XAI for computer vision.
May, 2024
· I had the honor to attend the workshop of the ELLIS Program in Machine Learning and Computer Vision and present our latest works on evaluating explainable AI methods.
April, 2024
· I will give a lecture on explainable AI together with Simone Schaub-Meyer.
April, 2024
· Our work "FunnyBirds" was features in the Computer Vision News under "BEST OF ICCV".
November, 2023
· Our paper "FunnyBirds" has been accepted at GCPR 2023 for an oral presentation in the Nectar Track.
August, 2023
· Our paper "FunnyBirds: A Synthetic Vision Dataset for a Part-Based Analysis of Explainable AI Methods" has been accepted at ICCV 2023 for an oral presentation.
July, 2023
· Our paper "Content-Adaptive Downsampling in Convolutional Neural Networks" has been accepted at CVPRW 2023.
April, 2023
· Our paper "Fast Axiomatic Attribution for Neural Networks" has been accepted at NeurIPS 2021.
September, 2021
· I will start a new position as a PhD student at Technical University of Darmstadt.
January, 2021

Publications

  • ...

    Benchmarking the Attribution Quality of Vision Models

    R. Hesse, S. Schaub-Meyer, S. Roth, in arXiv:2407.11910 [cs.CV], 2024.

    An evaluation protocol for attribution methods that overcomes the limitations of lacking inter-model comparisons and unaligned training and testing domains.

  • ...

    FunnyBirds: A Synthetic Vision Dataset for a Part-Based Analysis of Explainable AI Methods

    R. Hesse, S. Schaub-Meyer, S. Roth, in International Conference on Computer Vision (ICCV), 2023, oral presentation.

    A novel dataset that allows for image space interventions to evaluate explainable artificial intelligence (XAI) methods.

  • ...

    Content-Adaptive Downsampling in Convolutional Neural Networks

    R. Hesse, S. Schaub-Meyer, S. Roth, in CVPR 2023 Workshop on Efficient Deep Learning for Computer Vision (ECV).

    We propose an adaptive downsampling scheme that allows CNNs to process informative regions at a higher resolution than less informative ones.

  • ...

    Fast Axiomatic Attribution for Neural Networks

    R. Hesse, S. Schaub-Meyer, S. Roth, in Advances in Neural Information Processing Systems (NeurIPS), vol. 34, 2021.

    For ReLU DNNs without bias-terms, we show that Input×Gradient is a closed-form solution of Integrated Gradients that is therefore significantly more efficient to compute than the usual numerical approximation.

  • ...

    Homography Estimation in the Realm of Deep Learning

    Robin Hesse - Master's Thesis (2020)

    Even though deep learning has caused a paradigm shift in most computer vision applications, traditional handcrafted methods are still used extensively for image registration. In this thesis, we show that deep learning also caused a paradigm shift in image registration...

  • ...

    Development and Evaluation of 3D Autoencoders for Feature Extraction

    Robin Hesse - Bachelor's Thesis (2017)

    This thesis addresses the problem of aligning connectome data. The advancing scale in connectome reconstruction introduces new problems. One particular problem are spatial shifts and distortions in the image data that make it hard to come to a global alignment...

  • ...

    Efficient Verification of Program Fragments: Eager POR

    P. Metzler, H. Saissi, P. Bokor, R. Hesse, and N. Suri, in Automated Technology for Verification and Analysis, 2016.

    Software verification of concurrent programs is hampered by an exponentially growing state space due to non-deterministic process scheduling. Partial order reduction (POR)-based verification has proven to be a powerful technique to handle large state spaces...


Education

Technical University of Darmstadt

PhD candidate
2021 - present

Technical University of Darmstadt

Master of Science
Computer Science

GPA: 1.11 (with honors)

2018 - 2020

University of Arkansas

Custom and Short-Term Program
Computer Science

GPA: 4.0 / 4.0

2018

Technical University of Darmstadt

Bachelor of Science
Computer Science

GPA: 1.65

2014 - 2018

Experience

Research Assistant

Visual Inference Lab @ TU Darmstadt | Prof. Stefan Roth

My research focuses on explainable deep learning methods and how to evaluate them. Further, I assist in computer science classes (e.g., Computer Vision II and Computer Architectures), teach the Explainable Artificial Intelligence for Computer Vision lecture, and supervise multiple students (theses & projects).

2021 - present

Intern

Amazon.com, Inc. | Amazon Rekognition

In order to improve image alignment methods, I was working on a novel self-supervised end-to-end keypoint detection and description network. My implementation was released under an open source license allowing me to use it as the basis of my master’s thesis.

2019

Intern

Fraunhofer Institute for Computer Graphics Research | Prof. André Stork

My internship was about classifying 3D scans of highly reflective surfaces. In order to classify the scans, I trained a PointNet neural network that is invariant to permutations in the input, and hence, capable of working with point clouds.

2017 - 2018

Freelancer

Application Development

I worked freelance on different projects related to web and mobile applications.

2017 - 2019

Intern

Max Planck Institute for Brain Research | Prof. Moritz Helmstaedter

Within the context of my bachelor’s thesis in the field of deep learning and computational neuroscience, I interned at the department of connectomics (Prof. Helmstaedter, MPI Brain). In extension of the PointNet neural network, which can be used for point sets, I have developed an autoencoder that extracts features from 3D models of nerve tissue.

2017

Undergraduate Research Assistant

Dependable Systems and Software Department @ TU Darmstadt | Prof. Neeraj Suri

I have investigated and implemented algorithms to obtain and reduce the state space of concurrent programs. Additionally, I was involved in a project where a new algorithm was proposed that requires less overhead during state space exploration and shows an improvement in the runtime performance compared to existing algorithms (Metzler et al. 2016).

2015 - 2016

Intern

Intelligent Autonomous Systems Lab @ TU Darmstadt | Prof. Jan Peters

Together with a team of fellow students, I programmed a robot arm using Robot Operating System (ROS) to automatically scan three-dimensional objects. The scans were then classified and stored to build a knowledge base. I was highly involved in the registration process of the different views and in the classification of the objects.

2016