MVMO: A Multi-Object Dataset for Wide Baseline Multi-View Semantic Segmentation
This dataset addresses the challenge of occlusions in semantic segmentation for computer vision researchers, though it is incremental as it builds on existing multi-view datasets by adding specific features.
The authors tackled the problem of multi-view semantic segmentation by introducing MVMO, a synthetic dataset of 116,000 scenes with 10 object classes and 25 camera views, designed to feature wide baselines and high object density, which results in large disparities and occlusions.
We present MVMO (Multi-View, Multi-Object dataset): a synthetic dataset of 116,000 scenes containing randomly placed objects of 10 distinct classes and captured from 25 camera locations in the upper hemisphere. MVMO comprises photorealistic, path-traced image renders, together with semantic segmentation ground truth for every view. Unlike existing multi-view datasets, MVMO features wide baselines between cameras and high density of objects, which lead to large disparities, heavy occlusions and view-dependent object appearance. Single view semantic segmentation is hindered by self and inter-object occlusions that could benefit from additional viewpoints. Therefore, we expect that MVMO will propel research in multi-view semantic segmentation and cross-view semantic transfer. We also provide baselines that show that new research is needed in such fields to exploit the complementary information of multi-view setups.