paper-with-me

Cityscapes

홈페이지 · 논문 3,702편

Cityscapes is a large-scale database which focuses on semantic understanding of urban street scenes. It provides semantic, instance-wise, and dense pixel annotations for 30 classes grouped into 8 categories (flat surfaces, humans, vehicles, constructions, objects, nature, sky, and void). The dataset consists of around 5000 fine annotated images and 20000 coarse annotated ones. Data was captured in 50 cities during several months, daytimes, and good weather conditions. It was originally recorded as video so the frames were manually selected to have the following features: large number of dynamic objects, varying scene layout, and varying background. Source: A Review on Deep Learning Techniques Applied to Semantic Segmentation Image Source: https://www.cityscapes-dataset.com/dataset-overview/

Images

벤치마크

Semantic Segmentation on Cityscapes test 결과 211개
Semantic Segmentation on Cityscapes val 결과 198개
Panoptic Segmentation on Cityscapes val 결과 111개
Real-Time Semantic Segmentation on Cityscapes test 결과 78개
Semi-Supervised Semantic Segmentation on Cityscapes 12.5% labeled 결과 66개
Robust Object Detection on Cityscapes 결과 65개
Image-to-Image Translation on Cityscapes Labels-to-Photo 결과 63개
Semi-Supervised Semantic Segmentation on Cityscapes 25% labeled 결과 60개
Real-Time Semantic Segmentation on Cityscapes val 결과 48개
Semi-Supervised Semantic Segmentation on Cityscapes 50% labeled 결과 46개
Unsupervised Semantic Segmentation on Cityscapes test 결과 42개
Semi-Supervised Semantic Segmentation on Cityscapes 6.25% labeled 결과 36개
Unsupervised Semantic Segmentation with Language-image Pre-training on Cityscapes val 결과 36개
Video Semantic Segmentation on Cityscapes val 결과 36개
Panoptic Segmentation on Cityscapes test 결과 30개
Semi-Supervised Semantic Segmentation on Cityscapes 100 samples labeled 결과 26개
Semantic Segmentation on Cityscapes 결과 21개
Instance Segmentation on Cityscapes val 결과 17개
Image-to-Image Translation on Cityscapes Photo-to-Labels 결과 15개
Instance Segmentation on Cityscapes test 결과 11개
Robust Object Detection on Cityscapes test 결과 10개
Unsupervised Panoptic Segmentation on Cityscapes 결과 10개
Video Prediction on Cityscapes 128x128 결과 10개
Federated Learning on Cityscapes heterogeneous 결과 9개
Image Generation on Cityscapes 결과 6개
Monocular Depth Estimation on Cityscapes 결과 6개
Multi-Task Learning on Cityscapes test 결과 6개
Open Vocabulary Semantic Segmentation on Cityscapes 결과 6개
Semi-Supervised Semantic Segmentation on Cityscapes 2% labeled 결과 6개
Semi-Supervised Semantic Segmentation on Cityscapes 5% labeled 결과 6개
Semi-Supervised Semantic Segmentation on Cityscapes 93 labeled 결과 6개
Video Prediction on Cityscapes 결과 6개
Overlapped 10-1 on Cityscapes 결과 5개
Overlapped 14-1 on Cityscapes 결과 5개
Depth Estimation on Cityscapes test 결과 4개
Edge Detection on Cityscapes test 결과 4개
Real-Time Semantic Segmentation on Cityscapes 결과 4개
Semi-Supervised Semantic Segmentation on Cityscapes with extra (no coarse labels) 결과 4개
Unsupervised Semantic Segmentation on Cityscapes val 결과 3개
Scene Parsing on Cityscapes test 결과 2개
Semi-Supervised Semantic Segmentation on Cityscapes 10% labeled 결과 2개
Weakly-Supervised Semantic Segmentation on Cityscapes test 결과 2개
Weakly-Supervised Semantic Segmentation on Cityscapes val 결과 2개
2D Semantic Segmentation on Cityscapes val 결과 1개
Image Generation on Cityscapes-25K 256x512 결과 1개
Image Generation on Cityscapes-5K 256x512 결과 1개
Instance Segmentation on Cityscapes 결과 1개
Interactive Segmentation on Cityscapes val 결과 1개
Knowledge Distillation on Cityscapes 결과 1개
Real-time Instance Segmentation on Cityscapes test 결과 1개
Semi-Supervised Instance Segmentation on Cityscapes 결과 1개