Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in [1].
Event-based Action Recognition Using Motion Information and Spiking Neural Networks
Authors: Qianhui Liu, Dong Xing, Huajin Tang, De Ma, Gang Pan
IJCAI 2021 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We evaluate our proposed SNN on three event-based action recognition datasets, including our newly published Daily Action-DVS dataset comprising 12 actions collected under diverse recording conditions. Extensive experimental results show the effectiveness of motion information and our proposed SNN architecture for event-based action recognition. In this section, we evaluate the performance of our proposed SNN on three event-based gesture/action recognition datasets and compare it with other SNN methods. |
| Researcher Affiliation | Academia | 1College of Computer Science and Technology, Zhejiang University, Hangzhou, China 2Zhejiang Lab, Hangzhou, China EMAIL |
| Pseudocode | No | The paper describes the SNN architecture and its components using text and mathematical equations, but does not include any explicit pseudocode or algorithm blocks. |
| Open Source Code | Yes | 1https://github.com/qianhuiliu/SNN-action-recognition (The footnote is placed after the description of the Daily Action-DVS dataset, but the repository name "SNN-action-recognition" suggests it also contains the code for the proposed SNN.) |
| Open Datasets | Yes | Daily Action-DVS dataset: It comprises 1440 recordings of 15 subjects acting 12 different actions... 1https://github.com/qianhuiliu/SNN-action-recognition and publicly available Dvs Gesture dataset and Action Recognition dataset. |
| Dataset Splits | No | We use the full recordings (within 6000ms) for training and observe the performance of each method within the first 1000ms recordings. |
| Hardware Specification | No | No specific hardware details (like CPU/GPU models, memory) used for running experiments were mentioned in the paper. |
| Software Dependencies | No | No specific software dependencies with version numbers (e.g., Python, PyTorch, TensorFlow versions) were mentioned in the paper. |
| Experiment Setup | No | The paper describes the SNN architecture, the LIF model, and the SPA learning algorithm, and mentions 'λ is the learning rate', but it does not provide specific hyperparameter values such as the learning rate value itself, batch size, number of epochs, or detailed optimizer settings. |