Research map / Architectures and efficiency / CNN design
DeepCaps: Going Deeper With Capsule Networks
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use the deeper DeepCaps architecture instead of Sabour et al. CapsNet | Capsule Networks · CIFAR10 | accuracy: 89.40% -> 91.01% | Helped |
| Use the deeper DeepCaps architecture instead of Sabour et al. CapsNet | Capsule Networks · SVHN | accuracy: 95.70% -> 97.16% | Helped |
| Use the deeper DeepCaps architecture instead of HitNet | Capsule Networks · Fashion-MNIST | accuracy: 92.30% -> 94.46% | Helped |
| Use DeepCaps instead of the best reported state-of-the-art baseline | Image classification · MNIST | accuracy: 99.79% -> 99.72% | Hurt |
| Use DeepCaps instead of Sabour et al. CapsNet | Capsule Networks · MNIST | accuracy: 99.75% -> 99.72% | Hurt |
| Use a weighted ensemble of 7 models rather than a single DeepCaps model | DeepCaps · CIFAR10 | accuracy: 91.01% -> 92.74% | Helped |
| Use a weighted ensemble of 7 models rather than a single DeepCaps model | DeepCaps · SVHN | accuracy: 97.16% -> 97.56% | Helped |
| Use a weighted ensemble of 7 models rather than a single DeepCaps model | DeepCaps · Fashion-MNIST | accuracy: 94.46% -> 94.73% | Helped |
| Use DeepCaps instead of CapsNet | Capsule Networks on CIFAR10 · CIFAR10 | number of parameters: 22.48 Million -> 7.22 Million | Helped |
| Use DeepCaps instead of CapsNet | Capsule Networks · inference on NVIDIA V100 GPU | inference time: 2.86 ms -> 1.38 ms | Helped |
Closest papers by meaning
- STAR-Caps: Capsule Networks with Straight-Through Attentive Routing (NeurIPS 2019 · similarity 0.89)
- OrthCaps: An Orthogonal CapsNet with Sparse Attention Routing and Pruning (CVPR 2024 · similarity 0.88)
- CapProNet: Deep Feature Learning via Orthogonal Projections onto Capsule Subspaces (NeurIPS 2018 · similarity 0.88)
- Self-Routing Capsule Networks (NeurIPS 2019 · similarity 0.88)
- SubSpace Capsule Network (AAAI 2020 · similarity 0.87)
- ParseCaps: An Interpretable Parsing Capsule Network for Medical Image Diagnosis (AAAI 2025 · similarity 0.84)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.