Research map / Detection and segmentation / CNN features
Dilated Residual 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 |
|---|---|---|---|
| Replace late striding with dilated convolutions to retain higher-resolution features | ResNet-18 · ImageNet 2012 validation, 10-crop | top-1 error: 28.22 -> 25.75 | Helped |
| Replace late striding with dilated convolutions to retain higher-resolution features | ResNet-34 · ImageNet 2012 validation, 10-crop | top-1 error: 24.76 -> 22.64 | Helped |
| Add degridding changes: replace early max pooling, add lower-dilation blocks, and remove their residual connections (ablation) | DRN-A-18 · ImageNet 2012 validation, 10-crop | top-1 error: 25.75 -> 22.93 | Helped |
| Remove residual connections from the added lower-dilation blocks (ablation) | DRN-B-26 · ImageNet 2012 validation, 10-crop | top-1 error: 23.33 -> 22.93 | Helped |
| Add degridding changes (ablation) | DRN-A-34 · ImageNet 2012 validation, 10-crop | top-1 error: 22.64 -> 21.20 | Helped |
| Add the full degridding construction (ablation) | DRN-A-50 · Cityscapes validation | mean IoU: 67.3 -> 68.0 | Helped |
Closest papers by meaning
- Deep Residual Learning for Image Recognition (CVPR 2016 · similarity 0.90)
- Deep Pyramidal Residual Networks (CVPR 2017 · similarity 0.89)
- Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi-Supervised Semantic Segmentation (CVPR 2018 · similarity 0.88)
- Glance and Focus: a Dynamic Approach to Reducing Spatial Redundancy in Image Classification (NeurIPS 2020 · similarity 0.87)
- Learning Strict Identity Mappings in Deep Residual Networks (CVPR 2018 · similarity 0.86)
- Dr2Net: Dynamic Reversible Dual-Residual Networks for Memory-Efficient Finetuning (CVPR 2024 · similarity 0.86)
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.