# 3D ShapeNets: A Deep Representation for Volumetric Shapes

arXiv: https://arxiv.org/abs/1406.5670 · HTML: https://mlsanity.com/paper/1406.5670/

## Key facts

- Cited by 27 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 22 Jun 2014.

## Built on by (most cited first, 27 of 27)

- [RoboEXP: Action-Conditioned Scene Graph via Interactive Exploration for Robotic Manipulation](https://mlsanity.com/paper/2402.15487/) (2024)
- [PVT: Point-Voxel Transformer for Point Cloud Learning](https://mlsanity.com/paper/2108.06076/) (2021)
- [PatchFormer: An Efficient Point Transformer with Patch Attention](https://mlsanity.com/paper/2111.00207/) (2021)
- [SAM-RL: Sensing-Aware Model-Based Reinforcement Learning via Differentiable Physics-Based Simulation and Rendering](https://mlsanity.com/paper/2210.15185/) (2022)
- [A Unified Game-Theoretic Interpretation of Adversarial Robustness](https://mlsanity.com/paper/2111.03536/) (2021)
- [DSPoint: Dual-scale Point Cloud Recognition with High-frequency Fusion](https://mlsanity.com/paper/2111.10332/) (2021)
- [CloudFixer: Test-Time Adaptation for 3D Point Clouds via Diffusion-Guided Geometric Transformation](https://mlsanity.com/paper/2407.16193/) (2024)
- [FatNet: A Feature-attentive Network for 3D Point Cloud Processing](https://mlsanity.com/paper/2104.03427/) (2021)
- [GPr-Net: Geometric Prototypical Network for Point Cloud Few-Shot Learning](https://mlsanity.com/paper/2304.06007/) (2023)
- [Active Open-Vocabulary Recognition: Let Intelligent Moving Mitigate CLIP Limitations](https://mlsanity.com/paper/2311.17938/) (2023)
- [Learning Body-Aware 3D Shape Generative Models](https://mlsanity.com/paper/2112.07022/) (2021)
- [3D Face Morphing Attacks: Generation, Vulnerability and Detection](https://mlsanity.com/paper/2201.03454/) (2022)
- [Learning the Next Best View for 3D Point Clouds via Topological Features](https://mlsanity.com/paper/2103.02789/) (2021)
- [Explaining Deep Neural Networks for Point Clouds using Gradient-based Visualisations](https://mlsanity.com/paper/2207.12984/) (2022)
- [MVIP -- A Dataset and Methods for Application Oriented Multi-View and Multi-Modal Industrial Part Recognition](https://mlsanity.com/paper/2502.15448/) (2025)
- [Visual Attention in Imaginative Agents](https://mlsanity.com/paper/2104.00177/) (2021)
- [Bridging the Gap: Point Clouds for Merging Neurons in Connectomics](https://mlsanity.com/paper/2112.02039/) (2021)
- [Neuroplastic graph attention networks for nuclei segmentation in histopathology images](https://mlsanity.com/paper/2201.03669/) (2022)
- [PointTree: Transformation-Robust Point Cloud Encoder with Relaxed K-D Trees](https://mlsanity.com/paper/2208.05962/) (2022)
- [Semantic Segmentation on 3D Point Clouds with High Density Variations](https://mlsanity.com/paper/2307.01489/) (2023)
- [A Unified Game-Theoretic Interpretation of Adversarial Robustness](https://mlsanity.com/paper/2103.07364/) (2021)
- [Next-Best-View Estimation based on Deep Reinforcement Learning for Active Object Classification](https://mlsanity.com/paper/2110.06766/) (2021)
- [Learning task-specific features for 3D pointcloud graph creation](https://mlsanity.com/paper/2209.00949/) (2022)
- [Surface-biased Multi-Level Context 3D Object Detection](https://mlsanity.com/paper/2302.06291/) (2023)
- [Mahalanobis k-NN: A Statistical Lens for Robust Point-Cloud Registrations](https://mlsanity.com/paper/2409.06267/) (2024)
- [MorphoSkel3D: Morphological Skeletonization of 3D Point Clouds for Informed Sampling in Object Classification and Retrieval](https://mlsanity.com/paper/2501.12974/) (2025)
- [Denoising Diffusion Probabilistic Model for Point Cloud Compression at Low Bit-Rates](https://mlsanity.com/paper/2505.13316/) (2025)

## Abstract (arXiv)

3D shape is a crucial but heavily underutilized cue in today's computer vision systems, mostly due to the lack of a good generic shape representation. With the recent availability of inexpensive 2.5D depth sensors (e.g. Microsoft Kinect), it is becoming increasingly important to have a powerful 3D shape representation in the loop. Apart from category recognition, recovering full 3D shapes from view-based 2.5D depth maps is also a critical part of visual understanding. To this end, we propose to represent a geometric 3D shape as a probability distribution of binary variables on a 3D voxel grid, using a Convolutional Deep Belief Network. Our model, 3D ShapeNets, learns the distribution of complex 3D shapes across different object categories and arbitrary poses from raw CAD data, and discovers hierarchical compositional part representations automatically. It naturally supports joint object recognition and shape completion from 2.5D depth maps, and it enables active object recognition through view planning. To train our 3D deep learning model, we construct ModelNet -- a large-scale 3D CAD model dataset. Extensive experiments show that our 3D deep representation enables significant performance improvement over the-state-of-the-arts in a variety of tasks.