Transient Glimpses: Unveiling Occluded Backgrounds through the Spike Camera

Authors: Jiyuan Zhang, Shiyan Chen, Yajing Zheng, Zhaofei Yu, Tiejun Huang

AAAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results demonstrate the efficiency and generalization capabilities of our model in effectively removing dense occlusions across diverse scenes.
Researcher Affiliation Academia Jiyuan Zhang1, 2, Shiyan Chen1, 2, Yajing Zheng1, 2*, Zhaofei Yu1, 2, 3*, Tiejun Huang1, 2, 3 1School of Computer Science, Peking University 2National Key Laboratory for Multimedia Information Processing, Peking University 3Institute for Artificial Intelligence, Peking University {jyzhang,2301112005}@stu.pku.edu.cn, {yj.zheng,yuzf12,tjhuang}@pku.edu.cn
Pseudocode No The paper describes the architecture and process of Spk Occ Net with text and diagrams (Figure 4, 5, 7) but does not provide any formal pseudocode or algorithm blocks.
Open Source Code Yes Public project page: https://github.com/Leozhangjiyuan/Spike De Occlusion.
Open Datasets Yes Additionally, to facilitate research in occlusion removal, we introduce the S-OCC dataset, which consists of real-world spike-based data. Public project page: https://github.com/Leozhangjiyuan/Spike De Occlusion. Among these, 108 sequences are randomly picked for training, while the remaining 20 sequences are for testing.
Dataset Splits No The paper mentions 108 sequences for training and 20 for testing from the S-OCC dataset, but it does not specify a separate validation split or how validation was performed.
Hardware Specification No The paper does not provide any specific details about the hardware (e.g., GPU model, CPU type, memory) used to run the experiments. It only states that 'We train all networks with Py Torch'.
Software Dependencies No The paper mentions 'Py Torch' as the framework used for training networks, but it does not provide a specific version number for PyTorch or any other software dependencies.
Experiment Setup Yes We train all networks with Py Torch. L1 loss is used for optimization. Quantitative metrics are peak signal-to-noise ratio (PSNR) and structural similarity (SSIM).