Lune

CVPR2024Top-tier venue

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

Linglin Jing, Yiming Ding, Yunpeng Gao, Zhigang Wang, Xu Yan, Dong Wang, Gerald Schaefer, Hui Fang, Bin Zhao, Xuelong Li

Abstract

Event-based semantic segmentation has gained popularity due to its capability to deal with scenarios under highspeed 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 pseudolabeling. In this paper, we propose a novel hybrid pseudolabeling framework for unsupervised event-based semantic segmentation, HPL-ESS, to alleviate the influence of noisy pseudo labels. Specifically, we first employ a plain unsupervised domain adaptation framework as our baseline, which can generate a set of pseudo labels through selftraining. 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 (SPA) module to further improve the consistency of target domain features. Extensive experiments show that the proposed method outperforms existing state-of-the-art methods by a large margin on benchmarks (e.g., +5.88% accuracy, +10.32% mIoU on DSEC-Semantic dataset), and even surpasses several supervised methods.

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

9 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • EventDrive: Event Cameras for Vision-Language Driving Intelligence cites Talk2Event: Grounded Understanding of Dynamic Scenes from Event Cameras
  • ESEG: Event-Based Segmentation Boosted by Explicit Edge-Semantic Guidance cites SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
  • DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic Segmentation cites SegFormer: Simple and Efficient Design for Semantic Segmentation with Transformers
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers9

  1. ESEG: Event-Based Segmentation Boosted by Explicit Edge-Semantic GuidanceYucheng Zhao, Gengyu Lyu, Ke Li, Zihao Wang et al.AAAI 2025 · 8 citations
  2. Talk2Event: Grounded Understanding of Dynamic Scenes from Event CamerasLingdong Kong, Dongyue Lu, Alan Liang, Rong Li et al.NeurIPS 2025 · 7 citations
  3. Efficient Event-Based Semantic Segmentation via Exploiting Frame-Event Fusion: A Hybrid Neural Network ApproachHebei Li, Yansong Peng, Jiahui Yuan, Peixi Wu et al.AAAI 2025 · 3 citations
  4. Depth Any Event Stream: Enhancing Event-based Monocular Depth Estimation via Dense-to-Sparse DistillationJinjing Zhu, Tianbo Pan, Zidong Cao, Yexin Liu et al.ICCV 2025 · 3 citations
  5. EventDrive: Event Cameras for Vision-Language Driving IntelligenceDongyue Lu, Rong Li, Ao Liang, Lingdong Kong et al.CVPR 2026 · 2 citations

Builds on11

  1. SegFormer: Simple and Efficient Design for Semantic Segmentation with TransformersEnze Xie, Wenhai Wang, Zhiding Yu, Anima Anandkumar et al.NeurIPS 2021 · 9,661 citations
  2. DAFormer: Improving Network Architectures and Training Strategies for Domain-Adaptive Semantic SegmentationLukas Hoyer, Dengxin Dai, Luc Van GoolCVPR 2022 · 562 citations
  3. Exploiting Domain-Specific Features to Enhance Domain GeneralizationManh-Ha Bui, Toan Tran, Anh Tran, Dinh Q. PhungNeurIPS 2021 · 182 citations
  4. Generalized Jensen-Shannon Divergence Loss for Learning with Noisy LabelsErik Englesson, Hossein AzizpourNeurIPS 2021 · 170 citations
  5. Understanding Gradual Domain Adaptation: Improved Analysis, Optimal Path and BeyondHaoxiang Wang, Bo Li, Han ZhaoICML 2022 · 48 citations

Related papers

  1. EvDistill: Asynchronous Events To End-Task Learning via Bidirectional Reconstruction-Guided Cross-Modal Knowledge DistillationLin Wang, Yujeong Chae, Sung-Hoon Yoon, Tae-Kyun Kim et al.CVPR 2021
  2. Re-coding for Uncertainties: Edge-awareness Semantic Concordance for Resilient Event-RGB SegmentationNan Bao, Yifan Zhao, Lin Zhu, Jia LiNeurIPS 2025 · 1 citation
  3. From Softmax to Dirichlet: Evidential Learning for Semi-supervised Semantic SegmentationHuayu Mai, Rui Sun, Yujia Chen, Wangkai Li et al.CVPR 2026
  4. CMDA: Cross-Modality Domain Adaptation for Nighttime Semantic SegmentationRuihao Xia, Chaoqiang Zhao, Meng Zheng, Ziyan Wu et al.ICCV 2023 · 54 citations
  5. Learning Adaptive Dense Event Stereo from the Image DomainHoonhee Cho, Jegyeong Cho, Kuk-Jin YoonCVPR 2023
Dusk over the sea between two cliffs drawn in fine vertical lines