mlsanityData to 7 Oct 2026

ShapeAug++: More Realistic Shape Augmentation for Event Data

arXiv 2409.11075 ↗First posted 17 Sep 2024cs.CVcited by 0

Not on the map. Why?

Abstract (arXiv)

The novel Dynamic Vision Sensors (DVSs) gained a great amount of attention recently as they are superior compared to RGB cameras in terms of latency, dynamic range and energy consumption. This is particularly of interest for autonomous applications since event cameras are able to alleviate motion blur and allow for night vision. One challenge in real-world autonomous settings is occlusion where foreground objects hinder the view on traffic participants in the background. The ShapeAug method addresses this problem by using simulated events resulting from objects moving on linear paths for event data augmentation. However, the shapes and movements lack complexity, making the simulation fail to resemble the behavior of objects in the real world. Therefore in this paper, we propose ShapeAug++, an extended version of ShapeAug which involves randomly generated polygons as well as curved movements. We show the superiority of our method on multiple DVS classification datasets, improving the top-1 accuracy by up to 3.7% compared to ShapeAug.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 15 of its 24 references that are in the map. First posted on arXiv 17 Sep 2024.

Builds on

15 of 24 references.

References · 15

  1. 2019CutMix: Regularization Strategy to Train Strong Classifiers with Localizable Featurescited by 2,875
  2. 2016Identity Mappings in Deep Residual Networkscited by 1,960
  3. 2017Improved Regularization of Convolutional Neural Networks with Cutoutcited by 1,623
  4. 2020Incorporating Learnable Membrane Time Constant to Enhance Learning of Spiking Neural Networkscited by 340
  5. 2015Converting Static Image Datasets to Spiking Neuromorphic Datasets Using Saccadescited by 308
  6. 2018HATS: Histograms of Averaged Time Surfaces for Robust Event-based Object Classificationcited by 215
  7. 2017Hide-and-Seek: Forcing a Network to be Meticulous for Weakly-supervised Object and Action Localizationcited by 207
  8. 2022Recurrent Vision Transformers for Object Detection with Event Camerascited by 158
  9. 2020A Large Scale Event-based Detection Dataset for Automotivecited by 108
  10. 2022Neuromorphic Data Augmentation for Training Spiking Neural Networkscited by 52
  11. 2019Augment your batch: better training with larger batchescited by 37
  12. 2021EventDrop: data augmentation for event-based learningcited by 28
3 more
  1. 2024EventRPG: Event Data Augmentation with Relevance Propagation Guidancecited by 7
  2. 2024ShapeAug: Occlusion Augmentation for Event Camera Datacited by 6
  3. 2019Occlusions for Effective Data Augmentation in Image Classificationcited by 5

Built on by

Not cited yet.