Mesh Saliency via Weakly Supervised Classification-for-Saliency CNN

A Multi-Modal Attention-Based Approach for Points of Interest Detection on 3D Shapes

Identifying points of interest (POIs) on the surface of 3D shapes is a significant challenge in geometric processing research. The complex connection between POIs and their geometric descriptors, combined with the small percentage of POIs on the shape, makes detecting POIs on any given 3D shape a highly challenging task. Existing methods directly detect POIs from the entire 3D shape, resulting in low efficiency and accuracy. Therefore, we propose a novel multi-modal POI detection method using a coarse-to-fine approach, with the key idea of reducing data complexity and enabling more efficient and accurate subsequent POI detection by first identifying and processing important regions on the 3D shape. It first obtains important areas on the 3D shape through 2D projected images, then processes points within these regions using attention mechanisms. Extensive experiments demonstrate that our method outperforms existing POI detection techniques.

SAL3D: a model for saliency prediction in 3D meshes

Advances in virtual and augmented reality have increased the demand for immersive and engaging 3D experiences. To create such experiences, it is crucial to understand visual attention in 3D environments, which is typically modeled by means of saliency maps. While attention in 2D images and traditional media has been widely studied, there is still much to explore in 3D settings. In this work, we propose a deep learning-based model for predicting saliency when viewing 3D objects, which is a first step toward understanding and predicting attention in 3D environments. Previous approaches rely solely on low-level geometric cues or unnatural conditions, however, our model is trained on a dataset of real viewing data that we have manually captured, which indeed reflects actual human viewing behavior. Our approach outperforms existing state-of-the-art methods and closely approximates the ground-truth data. Our results demonstrate the effectiveness of our approach in predicting attention in 3D objects, which can pave the way for creating more immersive and engaging 3D experiences.

Learning on 3D Meshes With Laplacian Encoding and Pooling

3D models are commonly used in computer vision and graphics. With the wider availability of mesh data, an efficient and intrinsic deep learning approach to processing 3D meshes is in great need. Unlike images, 3D meshes have irregular connectivity, requiring careful design to capture relations in the data. To utilize the topology information while staying robust under different triangulations, we propose to encode mesh connectivity using Laplacian spectral analysis, along with mesh feature aggregation blocks (MFABs) that can split the surface domain into local pooling patches and aggregate global information amongst them. We build a mesh hierarchy from fine to coarse using Laplacian spectral clustering, which is flexible under isometric transformations. Inside the MFABs there are pooling layers to collect local information and multi-layer perceptrons to compute vertex features of increasing complexity. To obtain the relationships among different clusters, we introduce a Correlation Net to compute a correlation matrix, which can aggregate the features globally by matrix multiplication with cluster features. Our network architecture is flexible enough to be used on meshes with different numbers of vertices. We conduct several experiments including shape segmentation and classification, and our method outperforms state-of-the-art algorithms for these tasks on the ShapeNet and COSEG datasets.