Lossy Event Compression Based On Image-Derived Quad Trees And Poisson Disk Sampling. Banerjee, S., Wang, Z. W., Chopp, H. H., Cossairt, O., & Katsaggelos, A. K. In 2021 IEEE International Conference on Image Processing (ICIP), volume 2021-Septe, pages 2154–2158, sep, 2021. IEEE, IEEE.
Lossy Event Compression Based On Image-Derived Quad Trees And Poisson Disk Sampling [link]Paper  doi  abstract   bibtex   
Event cameras have provided new opportunities for tackling visual tasks under challenging scenarios over conventional RGB cameras. However, not much focus has been given on event compression algorithms. The main challenge for compressing events is its unique asynchronous form. To address this problem, we propose a novel event compression algorithm based on a quad tree (QT) segmentation map derived from the adjacent intensity images. The QT informs 2D spatial priority within the 3D space-time volume. In the event encoding step, events are first aggregated over time to form polarity-based event histograms. The histograms are then variably sampled via Poisson Disk Sampling prioritized by the QT based segmentation map. Next, differential encoding and run length encoding are employed for encoding the spatial and polarity information of the sampled events, respectively, followed by Huffman encoding to produce the final encoded events. Our algorithm achieves greater than 6× higher compression compared to the state of the art.
@inproceedings{banerjee2021lossy,
abstract = {Event cameras have provided new opportunities for tackling visual tasks under challenging scenarios over conventional RGB cameras. However, not much focus has been given on event compression algorithms. The main challenge for compressing events is its unique asynchronous form. To address this problem, we propose a novel event compression algorithm based on a quad tree (QT) segmentation map derived from the adjacent intensity images. The QT informs 2D spatial priority within the 3D space-time volume. In the event encoding step, events are first aggregated over time to form polarity-based event histograms. The histograms are then variably sampled via Poisson Disk Sampling prioritized by the QT based segmentation map. Next, differential encoding and run length encoding are employed for encoding the spatial and polarity information of the sampled events, respectively, followed by Huffman encoding to produce the final encoded events. Our algorithm achieves greater than 6× higher compression compared to the state of the art.},
archivePrefix = {arXiv},
arxivId = {2005.00974},
author = {Banerjee, Srutarshi and Wang, Zihao W. and Chopp, Henry H. and Cossairt, Oliver and Katsaggelos, Aggelos K.},
booktitle = {2021 IEEE International Conference on Image Processing (ICIP)},
doi = {10.1109/ICIP42928.2021.9506546},
eprint = {2005.00974},
isbn = {978-1-6654-4115-5},
issn = {15224880},
keywords = {Lossy event compression,Poisson disk sampling,Quad tree segmentation},
month = {sep},
organization = {IEEE},
pages = {2154--2158},
publisher = {IEEE},
title = {{Lossy Event Compression Based On Image-Derived Quad Trees And Poisson Disk Sampling}},
url = {https://ieeexplore.ieee.org/document/9506546/},
volume = {2021-Septe},
year = {2021}
}

Downloads: 0