paper-with-me

Trans10K

홈페이지 · 논문 33편

A large-scale dataset for transparent object segmentation, named Trans10K, consisting of 10,428 images of real scenarios with carefully manual annotations, which are 10 times larger than the existing datasets. Source: [Segmenting Transparent Objects in the Wild](/paper/segmenting-transparent-objects-in-the-wild)

벤치마크

Semantic Segmentation on Trans10K 결과 31개