arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1904.04474cs.CV
收藏

UG $^{2+}$途径2:评估和推进能见度差环境下图像理解的集体基准测试

UG$^{2+}$ Track 2: A Collective Benchmark Effort for Evaluating and Advancing Image Understanding in Poor Visibility Environments

Peking University(北京大学);Texas A&M University(得克萨斯农工大学);Chinese Academy of Sciences(中国科学院);University of Notre Dame(圣母大学)

Ye Yuan, Wenhan Yang, Wenqi Ren, Jiaying Liu, Walter J. Scheirer, Zhangyang Wang

摘要:

IEEE CVPR 2019中的UG $^{2+}$挑战赛旨在唤起对低级视觉技术如何在各种场景中有益于高级自动视觉识别的全面讨论和探索。在第二个轨道中,我们专注于在恶劣天气(雾霾,下雨)和低光照条件下造成的可见性增强中的物体或人脸检测。虽然现有的增强方法在经验上被期望帮助高级别的最终任务,但观察到在实践中并不总是如此。为了提供更彻底的检查和公平的比较,我们引入了三个基准集,分别在现实世界的朦胧,下雨和低光条件下收集,注释对象/面部注释。据我们所知,这是第一次也是目前最大的一次此类努力。通过级联现有的增强和检测模型的基线结果报告,表明我们的新数据的高度挑战性的性质,以及进一步的技术创新的大空间。我们期待来自广泛研究界的大量参与,共同应对这些挑战。

英文摘要:

The UG$^{2+}$ challenge in IEEE CVPR 2019 aims to evoke a comprehensive discussion and exploration about how low-level vision techniques can benefit the high-level automatic visual recognition in various scenarios. In its second track, we focus on object or face detection in poor visibility enhancements caused by bad weathers (haze, rain) and low light conditions. While existing enhancement methods are empirically expected to help the high-level end task, that is observed to not always be the case in practice. To provide a more thorough examination and fair comparison, we introduce three benchmark sets collected in real-world hazy, rainy, and low-light conditions, respectively, with annotate objects/faces annotated. To our best knowledge, this is the first and currently largest effort of its kind. Baseline results by cascading existing enhancement and detection models are reported, indicating the highly challenging nature of our new data as well as the large room for further technical innovations. We expect a large participation from the broad research community to address these challenges together.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑