Summary of Urbench: a Comprehensive Benchmark For Evaluating Large Multimodal Models in Multi-view Urban Scenarios, by Baichuan Zhou et al.
UrBench: A Comprehensive Benchmark for Evaluating Large Multimodal Models in Multi-View Urban Scenarios
by Baichuan Zhou, Haote Yang, Dairong Chen, Junyan Ye, Tianyi Bai, Jinhua Yu, Songyang Zhang, Dahua Lin, Conghui He, Weijia Li
First submitted to arxiv on: 30 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 The paper presents UrBench, a comprehensive benchmark for evaluating Large Multimodal Models (LMMs) in complex multi-view urban scenarios. The benchmark consists of 11.6K questions across four task dimensions: Geo-Localization, Scene Reasoning, Scene Understanding, and Object Understanding. LMMs struggle in these tasks, with even the best-performing GPT-4o lagging behind humans by an average performance gap of 17.4%. The results highlight inconsistencies in LMM behavior across different urban views. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary The paper creates a special set of questions to test how well computer models can understand cities. They made a big collection of questions that cover many different tasks, like finding where something is or what objects are in a picture. They tested 21 computer models and found out they’re not very good at understanding cities yet. Even the best one wasn’t as good as humans. |
Keywords
» Artificial intelligence » Gpt » Scene understanding
