Research map / Detection and segmentation / CNN features
Deep Co-Occurrence Feature Learning for Visual Object Recognition
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 |
|---|---|---|---|
| Add co-occurrence features to VGG-16 and concatenate them with its feature vector | VGG-16 · CUB200-2011 | accuracy: 70.4% -> 83.6% | Helped |
| Add co-occurrence layers to the last three building blocks | ResNet-152 · CUB200-2011 | accuracy: 73.3% -> 85.8% | Helped |
| Add co-occurrence features from the last three convolutional layers (ablation) | VGG-16 · CUB200-2011 | accuracy: 70.4% -> 82.6% | Helped |
| Increase the number of 1 × 1 filters from 32 to 64 to 128 (ablation) | Co-occurrence features with VGG-16 · CUB200-2011 | accuracy: 80.6 -> 82.0 -> 82.6 | Helped |
| Increase the number of 1 × 1 filters from 32 to 64 to 128 (ablation) | Co-occurrence features with ResNet-152 · CUB200-2011 | accuracy: 83.6 -> 85.3 -> 85.8 | Helped |
| Use co-occurrence vectors jointly from the last three layers rather than a single layer (ablation) | VGG-16 and ResNet-152 · CUB200-2011 | accuracy | Helped |
| Apply co-occurrence layers to earlier VGG-16 convolutional layers (ablation) | VGG-16 · CUB200-2011 | accuracy | Hurt |
| Use feature maps from the last five ResNet-152 building blocks (ablation) | ResNet-152 · CUB200-2011 | accuracy: 79.5% to 82.2% | No clear effect |
Closest papers by meaning
- Co-Occurrence Neural Network (CVPR 2019 · similarity 0.91)
- Multi-Part Object Representations via Graph Structures and Co-Part Discovery (NeurIPS 2026 · similarity 0.89)
- Feature Pyramid Networks for Object Detection (CVPR 2017 · similarity 0.89)
- A Comparative Analysis and Study of Multiview CNN Models for Joint Object Categorization and Pose Estimation (ICML 2016 · similarity 0.88)
- SPDA-CNN: Unifying Semantic Part Detection and Abstraction for Fine-Grained Recognition (CVPR 2016 · similarity 0.88)
- ObjectNet: A large-scale bias-controlled dataset for pushing the limits of object recognition models (NeurIPS 2019 · similarity 0.88)
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.