Extension of the official KITTI'15 dataset. independently moving instance segmentation ground truth to cover all moving objects, not just a selection of cars and vans. Instance Motion Segmentation of all moving objects Binary Motion Segmentation (background/foreground) Validation Masks Dataset contains: Instance Motion Segmenation for the training split of the KITTI
1 PAPER • NO BENCHMARKS YET
Please find more details of this dataset at https://alex-xun-xu.github.io/ProjectPage/CVPR_18/index.html 3D motion segmentation has been the key problem in computer vision research due to the application Traditional motion segmentation approaches are often evaluated on artificial dataset like Hopkins 155 [1] and its variants. We proposed the KT3DMoSeg to address the 3D motion segmentation problem in real world scenes. A Benchmark for the Comparison of 3-D Motion Segmentation Algorithms. CVPR, 2007. [2] A. Geiger, P. Lenz, C. Stiller, and R. Urtasun. Vision meets robotics: The kitti dataset.
1 PAPER • 1 BENCHMARK