SkillAgentSearch skills...

RandBox

[ICCV 2023] PyTorch implementation of RandBox

Install / Use

/learn @scuwyh2000/RandBox
About this skill

Quality Score

0/100

Supported Platforms

Universal

README

Random Boxes Are Open-world Object Detectors (ICCV 2023)

RandBox is a novel and effective model for open-world object detection.

Random Boxes Are Open-world Object Detectors
Yanghao Wang, Zhongqi Yue, Xian-sheng Hua, Hanwang Zhang
arXiv 2307.08249

Updates

  • (07/2023) Code is released.

Models

Task | K-mAP | U-R | WI | A-OSE | Download --- |:---:|:---:|:---:|:---:|:---: Task 1 | 61.8 | 10.6 | 0.0240 | 4498 |model Task 2 | 45.3 | 6.3 | 0.0078 | 1880 |model Task 3 | 39.4 | 7.8 | 0.0054 | 1452 |model Task 4 | 35.4 | - | - | - |model

Getting Started

The installation instruction and usage are in Getting Started with RandBox.

Citing RandBox

If you use RandBox in your research or wish to refer to the baseline results published here, please use the following BibTeX entry.

@InProceedings{Wang_2023_ICCV,
    author    = {Wang, Yanghao and Yue, Zhongqi and Hua, Xian-Sheng and Zhang, Hanwang},
    title     = {Random Boxes Are Open-world Object Detectors},
    booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
    month     = {October},
    year      = {2023},
    pages     = {6233-6243}
}
View on GitHub
GitHub Stars56
CategoryEducation
Updated7mo ago
Forks5

Languages

Python

Security Score

77/100

Audited on Aug 22, 2025

No findings