Summary of Hasper: An Image Repository For Hand Shadow Puppet Recognition, by Syed Rifat Raiyan et al.
HaSPeR: An Image Repository for Hand Shadow Puppet Recognition
by Syed Rifat Raiyan, Zibran Zarif Amio, Sabbir Ahmed
First submitted to arxiv on: 19 Aug 2024
Categories
- Main: Computer Vision and Pattern Recognition (cs.CV)
- Secondary: Artificial Intelligence (cs.AI)
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 A novel dataset, {}, is introduced to facilitate preservation of hand shadow puppetry, a dying art form. The dataset consists of 15,000 images across 15 classes extracted from professional and amateur clips. Pretrained image classification models are employed to establish baselines, showing skip-connected convolutional models outperforming attention-based transformer architectures. Lightweight models like MobileNetV2 perform well, making them suitable for mobile applications. A prototype application is created to explore using low-latency architectures in developing ombromanie teaching tools. The best-performing model, ResNet34, is analyzed through feature-spatial, explainability, and error analyses to gain insights into its decision-making process. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary Hand shadow puppetry is a dying art form that tells stories by projecting hand shadows onto flat surfaces. To help preserve it, researchers created a dataset of 15,000 images from professional and amateur performers. They used computer models to test how well different approaches worked. The results show that certain types of computer models are better than others at recognizing these shadow puppets. This information can be useful for developing tools to teach people about this art form. |
Keywords
» Artificial intelligence » Attention » Image classification » Transformer
