ScanNet200 3D Semantic instance benchmark
The 3D semantic instance prediction task involves detecting and segmenting the object in an 3D scan mesh.
Evaluation and metricsSimilarly to the ScanNet benchmark in ScanNet200 our evaluation ranks all methods according to the average precision for each class. We report the mean average precision AP at overlap 0.25 (AP 25%), overlap 0.5 (AP 50%), and over overlaps in the range [0.5:0.95:0.05] (AP) for all 200 categories. Note that multiple predictions of the same ground truth instance are penalized as false positives.
This table lists the benchmark results for the ScanNet200 3D semantic instance scenario.