Summary of Finding Dino: a Plug-and-play Framework For Zero-shot Detection Of Out-of-distribution Objects Using Prototypes, by Poulami Sinhamahapatra et al.
Finding Dino: A Plug-and-Play Framework for Zero-Shot Detection of Out-of-Distribution Objects Using Prototypes
by Poulami Sinhamahapatra, Franziska Schwaiger, Shirsha Bose, Huiyu Wang, Karsten Roscher, Stephan Guennemann
First submitted to arxiv on: 11 Apr 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 The proposed PRototype-based OOD detection Without Labels (PROWL) framework is a plug-and-play method for detecting and localizing unknown or out-of-distribution objects in various scenes, including safety-critical applications like autonomous vehicles. This inference-based approach relies on extracting relevant features from self-supervised pre-trained models and does not require training on the domain dataset. PROWL can be easily adapted to detect known classes in any operational design domain (ODD) without requiring additional labeled data. The framework achieves state-of-the-art results on road driving benchmarks, including SMIYC and Fishyscapes, and demonstrates generalisability to other domains such as rail and maritime. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary A new AI tool can help self-driving cars detect unknown objects in their path. This tool, called PROWL, doesn’t need to be trained on every possible scenario. Instead, it uses pre-trained models to figure out what’s important to look for. This makes it very useful for situations where there isn’t a lot of data available. PROWL has been tested and shown to work well in several different environments, including roads and railways. |
Keywords
» Artificial intelligence » Inference » Self supervised
