3D-R2N2: A Unified Approach for Single and Multi-view 3D Object Reconstruction

2 Apr 2016  ·  Christopher B. Choy, Danfei Xu, JunYoung Gwak, Kevin Chen, Silvio Savarese ·

Inspired by the recent success of methods that employ shape priors to achieve robust 3D reconstructions, we propose a novel recurrent neural network architecture that we call the 3D Recurrent Reconstruction Neural Network (3D-R2N2). The network learns a mapping from images of objects to their underlying 3D shapes from a large collection of synthetic data... Our network takes in one or more images of an object instance from arbitrary viewpoints and outputs a reconstruction of the object in the form of a 3D occupancy grid. Unlike most of the previous works, our network does not require any image annotations or object class labels for training or testing. Our extensive experimental analysis shows that our reconstruction framework i) outperforms the state-of-the-art methods for single view reconstruction, and ii) enables the 3D reconstruction of objects in situations when traditional SFM/SLAM methods fail (because of lack of texture and/or wide baseline). read more

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
3D Reconstruction Data3D−R2N2 3D-R2N2 3DIoU 0.560 # 4
3D Object Reconstruction Data3D−R2N2 3D-R2N2 Avg F1 39.01 # 5
3DIoU 0.56 # 7

Methods


No methods listed for this paper. Add relevant methods here