Research map / Architectures and efficiency / CNN design
Meta-Learning Convolutional Neural Architectures for Multi-Target Concrete Defect Classification With the COncrete DEfect BRidge IMage Dataset
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 |
|---|---|---|---|
| search task-specific CNN architectures with MetaQNN or ENAS | CODEBRIM multi-target defect classification · CODEBRIM test | multi-target accuracy: 70.77 -> 72.19 | Helped |
| search task-specific CNN architectures with MetaQNN or ENAS | CODEBRIM multi-target defect classification · CODEBRIM | parameters (M): 11.50 -> 2.88 | Helped |
| use ImageNet- or MINC-pretrained convolutional features instead of training features for CODEBRIM (ablation) | fine-tuned CNN baselines · CODEBRIM multi-target classification | multi-target accuracy | Hurt |
| use multi-target accuracy rather than average per-class accuracy as the architecture-search reward | MetaQNN and ENAS · CODEBRIM | multi-target accuracy | Helped |
| use MetaQNN rather than ENAS for architecture search | CODEBRIM · CODEBRIM multi-target classification | multi-target accuracy | No clear effect |
| use a learning-rate cycle between [10^-2, 10^-5] and batch size 16 (ablation) | T-CNN and VGG-A training · CODEBRIM validation | multi-target accuracy | Helped |
| increase patch size beyond 224 (ablation) | T-CNN · CODEBRIM | multi-target accuracy | Helped |
| use patches smaller than 224 (ablation) | T-CNN · CODEBRIM | multi-target accuracy | Hurt |
Closest papers by meaning
- Cracks in the Foundation: A Civil Infrastructure Dataset to Challenge Vision Foundation Models (NeurIPS 2026 · similarity 0.84)
- Attack-inspired Calibration Loss for Calibrating Crack Recognition (AAAI 2025 · similarity 0.81)
- MVREC: A General Few-shot Defect Classification Model Using Multi-View Region-Context (AAAI 2025 · similarity 0.81)
- Towards Open-Vocabulary Industrial Defect Understanding with a Large-Scale Multimodal Dataset (CVPR 2026 · similarity 0.80)
- Wavelet and Prototype Augmented Query-based Transformer for Pixel-level Surface Defect Detection (CVPR 2025 · similarity 0.80)
- SCSegamba: Lightweight Structure-Aware Vision Mamba for Crack Segmentation in Structures (CVPR 2025 · similarity 0.79)
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.