Summary of Snn-par: Energy Efficient Pedestrian Attribute Recognition Via Spiking Neural Networks, by Haiyang Wang et al.
SNN-PAR: Energy Efficient Pedestrian Attribute Recognition via Spiking Neural Networks
by Haiyang Wang, Qian Zhu, Mowen She, Yabo Li, Haoyu Song, Minghe Xu, Xiao Wang
First submitted to arxiv on: 10 Oct 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI); Neural and Evolutionary Computing (cs.NE)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The proposed Spiking Neural Network (SNN) based framework for energy-efficient Pedestrian Attribute Recognition (PAR) addresses the high energy consumption issue in recent years’ advancements. The framework comprises a spiking tokenizer module, transforming pedestrian images into spiking feature representations, followed by spiking Transformer backbone networks for feature extraction. Enhanced features are fed into feed-forward networks for attribute recognition, leveraging binary cross-entropy loss and knowledge distillation from artificial neural networks to the SNN. Extensive experiments on three PAR benchmark datasets validate the framework’s effectiveness. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper proposes a way to make pedestrian attribute recognition more energy-efficient using spiking neural networks. This is important because current methods use too much energy. The new method works by first turning images of pedestrians into special feature representations, then using these features to recognize attributes like height or clothing type. The results show that this approach is better than previous ones. |
Keywords
» Artificial intelligence » Cross entropy » Feature extraction » Knowledge distillation » Neural network » Tokenizer » Transformer
