Shelf-Supervised Mesh Prediction in the Wild
This addresses the challenge of 3D reconstruction from images for various objects in real-world settings, though it builds incrementally on existing methods by using off-the-shelf segmentation.
The paper tackles the problem of inferring 3D shape and pose from a single image using a learning-based approach trained on unstructured image collections with only segmentation outputs as supervision, achieving scalability on 50 categories in the wild, which is an order of magnitude more classes than prior work.
We aim to infer 3D shape and pose of object from a single image and propose a learning-based approach that can train from unstructured image collections, supervised by only segmentation outputs from off-the-shelf recognition systems (i.e. 'shelf-supervised'). We first infer a volumetric representation in a canonical frame, along with the camera pose. We enforce the representation geometrically consistent with both appearance and masks, and also that the synthesized novel views are indistinguishable from image collections. The coarse volumetric prediction is then converted to a mesh-based representation, which is further refined in the predicted camera frame. These two steps allow both shape-pose factorization from image collections and per-instance reconstruction in finer details. We examine the method on both synthetic and real-world datasets and demonstrate its scalability on 50 categories in the wild, an order of magnitude more classes than existing works.