Improved Panoramic Representation via Bidirectional Recurrent View Aggregation for 3D Model Retrieval.

In view-based 3D model retrieval task, extracting discriminative high-level features of models from projected images is considered an effective approach. The challenge of view-based 3D shape retrieval is that shape information of each view is limited due to information deficiency in projection. Traditional methods in this direction mostly convert the model into a panoramic view, making it hard to recognize the original shape. To resolve this problem, we propose a novel deep neural network, Recurrent Panorama Network (RePanoNet), which can learn to build panoramic representation from view sequences. A view sequence is rendered at a circle around the model to provide enough panoramic information. For each view sequence, we employ the bidirectional LSTM in RePanoNet to recognize spatial correlations between adjacent views to construct a panoramic feature. In our experiments on ModelNet and ShapeNet Core55, RePanoNet outperforms the methods in the State-of-the-Art, which demonstrates its effectiveness.