3D Shape Completion with Test-Time Training
This work addresses shape completion for 3D models, which is an incremental improvement over existing methods by reducing boundary artifacts through test-time training.
The paper tackles 3D shape completion by separately predicting fractured and restored parts with interconnected predictions, using a decoder network based on signed distance functions and test-time training. It shows significant improvements in chamfer distances across eight ShapeNet categories compared to previous methods that often produce artifacts.
This work addresses the problem of \textit{shape completion}, i.e., the task of restoring incomplete shapes by predicting their missing parts. While previous works have often predicted the fractured and restored shape in one step, we approach the task by separately predicting the fractured and newly restored parts, but ensuring these predictions are interconnected. We use a decoder network motivated by related work on the prediction of signed distance functions (DeepSDF). In particular, our representation allows us to consider test-time-training, i.e., finetuning network parameters to match the given incomplete shape more accurately during inference. While previous works often have difficulties with artifacts around the fracture boundary, we demonstrate that our overfitting to the fractured parts leads to significant improvements in the restoration of eight different shape categories of the ShapeNet data set in terms of their chamfer distances.