Research map / Architectures and efficiency / CNN design
Gated Convolutional Networks with Hybrid Connectivity for Image Classification
Chuanguang Yang, Zhulin An, Hui Zhu, Xiaolong Hu, Kun Zhang, Kaiqiang Xu, Chao Li, Yongjun Xu
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 HCGNet-A3 instead of the strongest listed baseline | CIFAR-10 image classification · CIFAR-10 test | error rate (%): 2.98 -> 2.14 | Improved |
| Use HCGNet-A3 instead of NASNet-A | CIFAR-100 image classification · CIFAR-100 test | error rate (%): 16.03 -> 15.96 | Improved |
| Use HCGNet-C instead of AOGNet-40M | ImageNet classification · ImageNet 2012 validation | Top-1 error rate (%): 19.8 -> 19.5 | Improved |
| Use HCGNet-C instead of AOGNet-40M | ImageNet classification · ImageNet 2012 validation | Top-5 error rate (%): 4.9 -> 4.8 | Improved |
| Use HCGNet-B-FPN instead of ResNet-50-FPN | Mask-RCNN backbone · COCO val2017 detection | bounding-box AP50: 59.9 -> 60.6 | Improved |
| Use HCGNet-B-FPN instead of AOGNet-12M-FPN | Mask-RCNN backbone · COCO val2017 detection | bounding-box AP50:95: 38.0 -> 38.3 | Improved |
| Use HCGNet-B-FPN instead of AOGNet-12M-FPN | Mask-RCNN backbone · COCO val2017 instance segmentation | mask AP50:95: 34.6 -> 35.2 | Improved |
Closest papers by meaning
- DenseNets Reloaded: Paradigm Shift Beyond ResNets and ViTs (ECCV 2024 · similarity 0.87)
- Convolutional Networks with Dense Connectivity (TPAMI 2019 · similarity 0.86)
- Densely Connected Convolutional Networks (CVPR 2017 · similarity 0.86)
- Multi-Scale Dense Networks for Resource Efficient Image Classification (ICLR 2018 · similarity 0.86)
- MicroNet: Improving Image Recognition With Extremely Low FLOPs (ICCV 2021 · similarity 0.85)
- PICNN: A Pathway towards Interpretable Convolutional Neural Networks (AAAI 2024 · 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.
