USB: Universal-Scale Object Detection Benchmark
arXiv:2103.14027
Abstract
Benchmarks, such as COCO, play a crucial role in object detection. However, existing benchmarks are insufficient in scale variation, and their protocols are inadequate for fair comparison. In this paper, we introduce the Universal-Scale object detection Benchmark (USB). USB has variations in object scales and image domains by incorporating COCO with the recently proposed Waymo Open Dataset and Manga109-s dataset. To enable fair comparison and inclusive research, we propose training and evaluation protocols. They have multiple divisions for training epochs and evaluation image resolutions, like weight classes in sports, and compatibility across training protocols, like the backward compatibility of the Universal Serial Bus. Specifically, we request participants to report results with not only higher protocols (longer training) but also lower protocols (shorter training). Using the proposed benchmark and protocols, we conducted extensive experiments using 15 methods and found weaknesses of existing COCO-biased methods. The code is available at https://github.com/shinya7y/UniverseNet .
BMVC 2022
References in corpus (11)
- YOLOv4: Optimal Speed and Accuracy of Object Detection
- Res2Net: A New Multi-scale Backbone Architecture
- Accurate, Large Minibatch SGD: Training ImageNet in 1 Hour
- DSSD : Deconvolutional Single Shot Detector
- MMDetection: Open MMLab Detection Toolbox and Benchmark
- The Open Images Dataset V4: Unified image classification, object detection, and visual relationship detection at scale
- DetectoRS: Detecting Objects with Recursive Feature Pyramid and Switchable Atrous Convolution
- Building a Manga Dataset "Manga109" with Annotations for Multimedia Applications
- Object Detection for Comics using Manga109 Annotations
- 1st Place Solutions of Waymo Open Dataset Challenge 2020 -- 2D Object Detection Track
- 2nd Place Solution for Waymo Open Dataset Challenge -- 2D Object Detection