EvSSC: Event-aided Semantic Scene Completion

IEEE ICASSP 2026

Shangwei Guo · Hao Shi · Song Wang · Xiaoting Yin · Kailun Yang · Kaiwei Wang

---

📺Qualitative Results

https://www.youtube.com/watch?v=C_8RtiH_HO0

---

🔨Installation (One‑click script)

We provide a bash script that sets up the complete environment including conda, PyTorch, mmcv‑full, mmdetection3d, and the deformable attention operators.

Prerequisites

• Linux OS (tested on Ubuntu 20.04)
• NVIDIA GPU with CUDA 11.1 and cuDNN ≥ 8
• Conda (Miniconda or Anaconda)

Steps

chmod +x install_env.sh
./install_env.sh

---

⏬Downloads

📥 SemanticKITTI-E – Download

📥 DSEC-SSC – Download

---

💻Train

Train EvSSC with with 4 GPUs

./tools/dist_train.sh ./projects/configs/voxformer/voxformer_mm-S_3D_event_ELM.py 4

---

⭕️TODO

• Release the code.
• Release the arXiv preprint.
• Release datasets.

---

☺️Acknowledgments

We thank the authors of the following open-source projects and datasets for their valuable contributions:

• VoxFormer – Sparse voxel transformer for camera-based 3D semantic scene completion.
• DSEC – Stereo event camera dataset for driving scenarios.
• SemanticKITTI – Large-scale LiDAR dataset for semantic scene understanding.

If you use these resources in your research, please consider citing them using the BibTeX entries below:

@InProceedings{Li_2023_CVPR,
    title     = {VoxFormer: Sparse Voxel Transformer for Camera-Based 3D Semantic Scene Completion},
    author    = {Li, Yiming and Yu, Zhiding and Choy, Christopher and Xiao, Chaowei and Alvarez, Jose M. and Fidler, Sanja and Feng, Chen and Anandkumar, Anima},
    booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)},
    month     = {June},
    year      = {2023},
    pages     = {9087-9098}
}

@article{dsec2021,
      title={DSEC: A Stereo Event Camera Dataset for Driving Scenarios},
      author={Gehrig, Mathias and Aarents, Willem and Gehrig, Daniel and Scaramuzza, Davide},
      journal={IEEE Robotics and Automation Letters},
      volume={6},
      number={3},
      pages={4947--4954},
      year={2021},
      publisher={IEEE}
}

@inproceedings{semantickitti2019,
      title={SemanticKITTI: A Dataset for Semantic Scene Understanding of LiDAR Sequences},
      author={Behley, Jens and Garbade, Martin and Milioto, Andres and Quenzel, Jan and Behnke, Sven and Stachniss, Cyrill and Gall, Juergen},
      booktitle={Proceedings of the IEEE/CVF International Conference on Computer Vision},
      pages={9297--9307},
      year={2019}
}

---

😄Citation

If our work is helpful to you, please consider citing us using the following BibTeX entry:

@inproceedings{guo2026evssc,
      title={Event-aided Semantic Scene Completion}, 
      author={Shangwei Guo and Hao Shi and Song Wang and Xiaoting Yin and Kailun Yang and Kaiwei Wang},
      booktitle={2026 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)},
      year={2026}
}