Publications

Please see my Google Scholar for a complete list.

2025

  1. ZeroCAD: Leveraging 2D Outpainting as Guidance for Zero-Shot 3D Instance-based Scene Reconstruction
    Driton Salihu, Adam Misik, Yuankai Wu, and 4 more authors
    In Under Review, 2025

2024

  1. DeepSPF: Spherical SO (3)-Equivariant Patches for Scan-to-CAD Estimation
    Driton Salihu, Adam Misik, Yuankai Wu, and 3 more authors
    In The Twelfth International Conference on Learning Representations (ICLR2024), 2024
    PDF
  2. PaintedScenes.png
    NPRF: Neural Painted Radiosity Fields for Neural Implicit Rendering and Surface Reconstruction
    Driton Salihu, Adam Misik, Yuankai Wu, and 2 more authors
    In ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
    DOI PDF
  3. HEGN: Hierarchical Equivariant Graph Neural Network for 9DoF Point Cloud Registration
    Adam Misik, Driton Salihu, Xin Su, and 2 more authors
    In 2024 IEEE International Conference on Robotics and Automation (ICRA), 2024
    DOI PDF

2023

  1. sgpcr_example.gif
    SGPCR: Spherical Gaussian Point Cloud Representation and its Application to Object Registration and Retrieval
    Driton Salihu, and Eckehard Steinbach
    In The 2023 IEEE/CVF Winter Conference on Applications of Computer Vision (WACV 2023), Jan 2023
    PDF
  2. COCCA: Point Cloud Completion through Cad Cross-Attention
    Adam Misik, Driton Salihu, Heike Brock, and 1 more author
    In 2023 IEEE International Conference on Image Processing (ICIP), Jan 2023
    DOI PDF
  3. Reconstruction with robustness: A semantic prior guided face super-resolution framework for multiple degradations
    Hongjun Wu, Haoran Qi, Huanrong Zhang, and 3 more authors
    Image and Vision Computing, Jan 2023
    PDF

2022

  1. OverviewStructure.png
    S2CMAF: Multi-Method Assessment Fusion for Scan-to-CAD Methods
    D. Salihu, A. Misik, M. Hofbauer, and 1 more author
    In 2022 IEEE International Symposium on Multimedia (ISM), Dec 2022
    DOI PDF
  2. Anaconga: Analytical hw-cnn co-design using nested genetic algorithms
    Nael Fasfous, Manoj Rohit Vemparala, Alexander Frickenstein, and 8 more authors
    In 2022 Design, Automation & Test in Europe Conference & Exhibition (DATE), Dec 2022
    PDF

2021

  1. Hardware-aware mixed-precision neural networks using in-train quantization
    Manoj Rohit Vemparala, Nael Fasfous, Lukas Frickenstein, and 6 more authors
    In British machine vision conference (BMVC), Dec 2021
    PDF
  2. Hw-flowq: A multi-abstraction level hw-cnn co-design quantization methodology
    Nael Fasfous, Manoj Rohit Vemparala, Alexander Frickenstein, and 7 more authors
    ACM Transactions on Embedded Computing Systems (TECS), Dec 2021
    PDF