Multispectral Video Semantic Segmentation: A Benchmark Dataset and Baseline

Wei Ji; Jingjing Li; Cheng Bian; Zongwei Zhou; Jiaying Zhao; Alan L. Yuille; Li Cheng

doi:10.1109/cvpr52729.2023.00112

CVPR2023顶会

Multispectral Video Semantic Segmentation: A Benchmark Dataset and Baseline

Wei Ji, Jingjing Li, Cheng Bian, Zongwei Zhou, Jiaying Zhao, Alan L. Yuille, Li Cheng

DOI 出版方

2023年份

15顶会引用

摘要

github.io/Multispectral-Video-Semantic-Segmentation KAIST_set03_V000_2.-2 RGBT234_man9 KAIST_set02_V004_1.-3 Figure 1. Multispectral Video Semantic Segmentation. Examples of three typical real-life video sequences under diverse conditions are given (daytime (left), nighttime and overexposure (middle), rainy and low-light (right)), where RGB images, thermal infrared images, and their pixel-level semantic annotations are shown through the first to the third rows, respectively.