Challenge,Method pixellevel,"InternImage: Exploring Large-Scale Vision Foundation Models with Deformable Convolutions" "Class level" Metric,Average,road,sidewalk,building,wall,fence,pole,trafficlight,trafficsign,vegetation,terrain,sky,person,rider,car,truck,bus,train,motorcycle,bicycle IoU,86.0981,98.8583,88.8123,94.867,72.478,71.1928,75.417,80.8672,84.7038,94.517,75.513,96.285,90.0799,79.9006,96.7627,85.2668,95.525,92.6055,80.0189,82.1929 iIoU,73.6097,,,,,,,,,,,,78.7711,66.8356,90.787,66.8543,76.9166,69.2472,66.1576,73.3083 "Category level" Metric,Average,flat,nature,object,sky,construction,human,vehicle IoU,93.0483,98.8731,94.2884,80.4482,96.285,94.9866,90.0229,96.4336 iIoU,84.984,,,,,,79.7355,90.2324