# A Deeper Look at 3D Shape Classifiers

arXiv: https://arxiv.org/abs/1809.02560 · HTML: https://mlsanity.com/paper/1809.02560/

## Key facts

- Cited by 25 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 7 Sep 2018.

## Built on by (most cited first, 25 of 25)

- [PointCLIP V2: Prompting CLIP and GPT for Powerful 3D Open-world Learning](https://mlsanity.com/paper/2211.11682/) (2022)
- [Learning 3D Representations from 2D Pre-trained Models via Image-to-Point Masked Autoencoders](https://mlsanity.com/paper/2212.06785/) (2022)
- [SplitNets: Designing Neural Architectures for Efficient Distributed Computing on Head-Mounted Systems](https://mlsanity.com/paper/2204.04705/) (2022)
- [Learning Rotation-Invariant Representations of Point Clouds Using Aligned Edge Convolutional Neural Networks](https://mlsanity.com/paper/2101.00483/) (2021)
- [MV-CLIP: Multi-View CLIP for Zero-shot 3D Shape Recognition](https://mlsanity.com/paper/2311.18402/) (2023)
- [PointMCD: Boosting Deep Point Cloud Encoders via Multi-view Cross-modal Distillation for 3D Shape Recognition](https://mlsanity.com/paper/2207.03128/) (2022)
- [Improving Privacy-Preserving Vertical Federated Learning by Efficient Communication with ADMM](https://mlsanity.com/paper/2207.10226/) (2022)
- [MvDeCor: Multi-view Dense Correspondence Learning for Fine-grained 3D Segmentation](https://mlsanity.com/paper/2208.08580/) (2022)
- [Learning Canonical View Representation for 3D Shape Recognition with Arbitrary Views](https://mlsanity.com/paper/2108.07084/) (2021)
- [Invariance-Aware Randomized Smoothing Certificates](https://mlsanity.com/paper/2211.14207/) (2022)
- [AttWalk: Attentive Cross-Walks for Deep Mesh Analysis](https://mlsanity.com/paper/2104.11571/) (2021)
- [Enhancing Real-World Adversarial Patches through 3D Modeling of Complex Target Scenes](https://mlsanity.com/paper/2102.05334/) (2021)
- [SCA-PVNet: Self-and-Cross Attention Based Aggregation of Point Cloud and Multi-View for 3D Object Retrieval](https://mlsanity.com/paper/2307.10601/) (2023)
- [Unsupervised 3D Object Learning through Neuron Activity aware Plasticity](https://mlsanity.com/paper/2302.11622/) (2023)
- [VSFormer: Mining Correlations in Flexible View Set for Multi-view 3D Shape Understanding](https://mlsanity.com/paper/2409.09254/) (2024)
- [Edge-device Collaborative Computing for Multi-view Classification](https://mlsanity.com/paper/2409.15973/) (2024)
- [Spherical Transformer: Adapting Spherical Signal to CNNs](https://mlsanity.com/paper/2101.03848/) (2021)
- [Robust Pooling through the Data Mode](https://mlsanity.com/paper/2106.10850/) (2021)
- [PREMA: Part-based REcurrent Multi-view Aggregation Network for 3D Shape Retrieval](https://mlsanity.com/paper/2111.04945/) (2021)
- [ViewFormer: View Set Attention for Multi-view 3D Shape Understanding](https://mlsanity.com/paper/2305.00161/) (2023)
- [Automatic Reverse Engineering: Creating computer-aided design (CAD) models from multi-view images](https://mlsanity.com/paper/2309.13281/) (2023)
- [Beyond Viewpoint: Robust 3D Object Recognition under Arbitrary Views through Joint Multi-Part Representation](https://mlsanity.com/paper/2407.03842/) (2024)
- [Proto-FG3D: Prototype-based Interpretable Fine-Grained 3D Shape Classification](https://mlsanity.com/paper/2505.17666/) (2025)
- [A Geometric-Relational Deep Learning Framework for BIM Object Classification](https://mlsanity.com/paper/2212.00942/) (2022)
- [AegisRF: Adversarial Perturbations Guided with Sensitivity for Protecting Intellectual Property of Neural Radiance Fields](https://mlsanity.com/paper/2510.19371/) (2025)

## Abstract (arXiv)

We investigate the role of representations and architectures for classifying 3D shapes in terms of their computational efficiency, generalization, and robustness to adversarial transformations. By varying the number of training examples and employing cross-modal transfer learning we study the role of initialization of existing deep architectures for 3D shape classification. Our analysis shows that multiview methods continue to offer the best generalization even without pretraining on large labeled image datasets, and even when trained on simplified inputs such as binary silhouettes. Furthermore, the performance of voxel-based 3D convolutional networks and point-based architectures can be improved via cross-modal transfer from image representations. Finally, we analyze the robustness of 3D shape classifiers to adversarial transformations and present a novel approach for generating adversarial perturbations of a 3D shape for multiview classifiers using a differentiable renderer. We find that point-based networks are more robust to point position perturbations while voxel-based and multiview networks are easily fooled with the addition of imperceptible noise to the input.