KITTI-STEP
ImagesCC BY-NC-SA 3.0Introduced 2021-02-23
The Segmenting and Tracking Every Pixel (STEP) benchmark consists of 21 training sequences and 29 test sequences. It is based on the KITTI Tracking Evaluation and the Multi-Object Tracking and Segmentation (MOTS) benchmark. This benchmark extends the annotations to the Segmenting and Tracking Every Pixel (STEP) task. [Copy-pasted from http://www.cvlibs.net/datasets/kitti/eval_step.php]
Benchmarks
10-shot image generation/AQ10-shot image generation/SQ10-shot image generation/STQ10-shot image generation/mIoU10-shot image generation/pixel accuracyPanoptic Segmentation/AQPanoptic Segmentation/SQPanoptic Segmentation/STQSemantic Segmentation/AQSemantic Segmentation/SQSemantic Segmentation/STQSemantic Segmentation/mIoUSemantic Segmentation/pixel accuracyUnsupervised Semantic Segmentation/mIoUUnsupervised Semantic Segmentation/pixel accuracy