HPL-ESS: Hybrid Pseudo-Labeling for Unsupervised Event-based Semantic Segmentation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jing, Linglin, Ding, Yiming, Gao, Yunpeng, Wang, Zhigang, Yan, Xu, Wang, Dong, Schaefer, Gerald, Fang, Hui, Zhao, Bin, Li, Xuelong
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929287917993984
author Jing, Linglin
Ding, Yiming
Gao, Yunpeng
Wang, Zhigang
Yan, Xu
Wang, Dong
Schaefer, Gerald
Fang, Hui
Zhao, Bin
Li, Xuelong
author_facet Jing, Linglin
Ding, Yiming
Gao, Yunpeng
Wang, Zhigang
Yan, Xu
Wang, Dong
Schaefer, Gerald
Fang, Hui
Zhao, Bin
Li, Xuelong
contents Event-based semantic segmentation has gained popularity due to its capability to deal with scenarios under high-speed motion and extreme lighting conditions, which cannot be addressed by conventional RGB cameras. Since it is hard to annotate event data, previous approaches rely on event-to-image reconstruction to obtain pseudo labels for training. However, this will inevitably introduce noise, and learning from noisy pseudo labels, especially when generated from a single source, may reinforce the errors. This drawback is also called confirmation bias in pseudo-labeling. In this paper, we propose a novel hybrid pseudo-labeling framework for unsupervised event-based semantic segmentation, HPL-ESS, to alleviate the influence of noisy pseudo labels. In particular, we first employ a plain unsupervised domain adaptation framework as our baseline, which can generate a set of pseudo labels through self-training. Then, we incorporate offline event-to-image reconstruction into the framework, and obtain another set of pseudo labels by predicting segmentation maps on the reconstructed images. A noisy label learning strategy is designed to mix the two sets of pseudo labels and enhance the quality. Moreover, we propose a soft prototypical alignment module to further improve the consistency of target domain features. Extensive experiments show that our proposed method outperforms existing state-of-the-art methods by a large margin on the DSEC-Semantic dataset (+5.88% accuracy, +10.32% mIoU), which even surpasses several supervised methods.
format Preprint
id arxiv_https___arxiv_org_abs_2403_16788
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle HPL-ESS: Hybrid Pseudo-Labeling for Unsupervised Event-based Semantic Segmentation
Jing, Linglin
Ding, Yiming
Gao, Yunpeng
Wang, Zhigang
Yan, Xu
Wang, Dong
Schaefer, Gerald
Fang, Hui
Zhao, Bin
Li, Xuelong
Computer Vision and Pattern Recognition
Event-based semantic segmentation has gained popularity due to its capability to deal with scenarios under high-speed motion and extreme lighting conditions, which cannot be addressed by conventional RGB cameras. Since it is hard to annotate event data, previous approaches rely on event-to-image reconstruction to obtain pseudo labels for training. However, this will inevitably introduce noise, and learning from noisy pseudo labels, especially when generated from a single source, may reinforce the errors. This drawback is also called confirmation bias in pseudo-labeling. In this paper, we propose a novel hybrid pseudo-labeling framework for unsupervised event-based semantic segmentation, HPL-ESS, to alleviate the influence of noisy pseudo labels. In particular, we first employ a plain unsupervised domain adaptation framework as our baseline, which can generate a set of pseudo labels through self-training. Then, we incorporate offline event-to-image reconstruction into the framework, and obtain another set of pseudo labels by predicting segmentation maps on the reconstructed images. A noisy label learning strategy is designed to mix the two sets of pseudo labels and enhance the quality. Moreover, we propose a soft prototypical alignment module to further improve the consistency of target domain features. Extensive experiments show that our proposed method outperforms existing state-of-the-art methods by a large margin on the DSEC-Semantic dataset (+5.88% accuracy, +10.32% mIoU), which even surpasses several supervised methods.
title HPL-ESS: Hybrid Pseudo-Labeling for Unsupervised Event-based Semantic Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.16788