Implicit Shape Completion via Adversarial Shape Priors
This addresses the problem of 3D shape completion from incomplete scans for computer vision and graphics applications, representing an incremental improvement over existing neural implicit methods.
The paper tackles partial point cloud completion by combining a conditional Deep-SDF architecture with adversarial shape priors, achieving state-of-the-art performance on both man-made objects and deformable shapes like human bodies.
We present a novel neural implicit shape method for partial point cloud completion. To that end, we combine a conditional Deep-SDF architecture with learned, adversarial shape priors. More specifically, our network converts partial inputs into a global latent code and then recovers the full geometry via an implicit, signed distance generator. Additionally, we train a PointNet++ discriminator that impels the generator to produce plausible, globally consistent reconstructions. In that way, we effectively decouple the challenges of predicting shapes that are both realistic, i.e. imitate the training set's pose distribution, and accurate in the sense that they replicate the partial input observations. In our experiments, we demonstrate state-of-the-art performance for completing partial shapes, considering both man-made objects (e.g. airplanes, chairs, ...) and deformable shape categories (human bodies). Finally, we show that our adversarial training approach leads to visually plausible reconstructions that are highly consistent in recovering missing parts of a given object.