PPLNs: Parametric Piecewise Linear Networks for Event-Based Temporal Modeling and Beyond
Abstract
We present Parametric Piecewise Linear Networks (PPLNs) for temporal vision inference. Motivated by the neuromorphic principles that regulate biological neural behaviors, PPLNs are ideal for processing data captured by event cameras, which are built to simulate neural activities in the human retina. We discuss how to represent the membrane potential of an artificial neuron by a parametric piecewise linear function with learnable coefficients. This design echoes the idea of building deep models from learnable parametric functions recently popularized by Kolmogorov–Arnold Networks (KANs). Experiments demonstrate the state-of-the-art performance of PPLNs in event-based and image-based vision applications, including steering prediction, human pose estimation, and motion deblurring.
Cite
Text
Song et al. "PPLNs: Parametric Piecewise Linear Networks for Event-Based Temporal Modeling and Beyond." Neural Information Processing Systems, 2024. doi:10.52202/079017-2308Markdown
[Song et al. "PPLNs: Parametric Piecewise Linear Networks for Event-Based Temporal Modeling and Beyond." Neural Information Processing Systems, 2024.](https://mlanthology.org/neurips/2024/song2024neurips-pplns/) doi:10.52202/079017-2308BibTeX
@inproceedings{song2024neurips-pplns,
title = {{PPLNs: Parametric Piecewise Linear Networks for Event-Based Temporal Modeling and Beyond}},
author = {Song, Chen and Liang, Zhenxiao and Sun, Bo and Huang, Qixing},
booktitle = {Neural Information Processing Systems},
year = {2024},
doi = {10.52202/079017-2308},
url = {https://mlanthology.org/neurips/2024/song2024neurips-pplns/}
}