PointDCCNet: 3D Object Categorization Network using Point Cloud Decomposition

Katageri, Siddharth; Kulmi, Sameer; Tabib, Ramesh Ashok; Mudenagudi, Uma

doi:10.1109/cvprw53098.2021.00249

Cited by 5 publications

(3 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In the physical world, all objects are represented in 3dimensions and so can have different shapes from different viewpoints. Thus, to understand the importance of 3D object shape in categorization based on 2D images is a hot topic in computer vision (CV) [1,2,3]. The difficulty of extracting 3D shapes from multiple views of the object, involve several challenges among which multi-view data acquisition in largescale is substantial.…”

Section: Introductionmentioning

confidence: 99%

“…Among several concrete contributions in 3D object categorization, very few articles are on point based generalization for multi-view object shape categorization. To overcome this limitation, CV researchers proposed various approaches such as proposing new complex networks [3], new regularization [8], new data augmentation methods [9], etc. Data augmentation can considerably up-lifts the data size and enhances the learning capability of deep models, so used majorly in all tasks.…”

Section: Introductionmentioning

confidence: 99%

“…Data augmentation can considerably up-lifts the data size and enhances the learning capability of deep models, so used majorly in all tasks. Currently, the geometrical shape representation of 3dimensional objects are very flexible but unorganized [9,3]. Networks such as VoxNet [10] and its variants are incapable of fetching sufficient right geometry from the original 3D object.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Implicit Shape Biased Few-Shot Learning for 3D Object Generalization

Prasad

Lin

et al. 2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

The current state-of-the-art (SOTA) methods validate the role of shape in object categorization, however, except few, most of them neglect object shape information. Motivated by lowshot learning and increasing synthetic data in vision tasks, we investigated how image-based embedding generalization can be improved by the data itself. We propose a new data augmentation approach for low-shot object generalization regime based on image-only. The proposed method learns a discriminative embedding space using SIFT shape points for 3D objects, such that it's easier to map images and point clouds into one. Numerous experiments show that the proposed approach is superior to the existing low-shot SOTA methods.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Implicit Shape Biased Few-Shot Learning for 3D Object Generalization

Prasad

Lin

et al. 2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

show abstract

Variational autoencoders for 3D data processing

Molnár,

Tamás

2024

Artif Intell Rev

View full text Add to dashboard Cite

Variational autoencoders (VAEs) play an important role in high-dimensional data generation based on their ability to fuse the stochastic data representation with the power of recent deep learning techniques. The main advantages of these types of generators lie in their ability to encode the information with the possibility to decode and generalize new samples. This capability was heavily explored for 2D image processing; however, only limited research focuses on VAEs for 3D data processing. In this article, we provide a thorough review of the latest achievements in 3D data processing using VAEs. These 3D data types are mostly point clouds, meshes, and voxel grids, which are the focus of a wide range of applications, especially in robotics. First, we shortly present the basic autoencoder with the extensions towards the VAE with further subcategories relevant to discrete point cloud processing. Then, the 3D data specific VAEs are presented according to how they operate on spatial data. Finally, a few comprehensive table summarizing the methods, codes, and datasets as well as a citation map is presented for a better understanding of the VAEs applied to 3D data. The structure of the analyzed papers follows a taxonomy, which differentiates the algorithms according to their primary data types and application domains.

show abstract