Abstract
We introduce Amazon Berkeley Objects (ABO), a new large-scale datasetdesigned to help bridge the gap between real and virtual 3D worlds. ABOcontains product catalog images, metadata, and artist-created 3D models withcomplex geometries and physically-based materials that correspond to real,household objects. We derive challenging benchmarks that exploit the uniqueproperties of ABO and measure the current limits of the state-of-the-art onthree open problems for real-world 3D object understanding: single-view 3Dreconstruction, material estimation, and cross-domain multi-view objectretrieval.
Quick Read (beta)
loading the full paper ...