Research map / Label-efficient and robust learning / Recognition and long-tail
A Dataset Generation Framework for Evaluating Megapixel Image Classifiers & Their Explanations
Gautam Machiraju, Sylvia Plevritis, Parag Mallick
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 attention-based VGG-Att instead of VGG-19 | patch CNN · MISO-1 MD PPM | sensitivity: 0.500 -> 0.721 | Improved |
| use VGG-Att instead of VGG-19 | patch CNN · MISO-1 DSP PPM | sensitivity: 0.500 -> 0.999 | Improved |
| use VGG-Att instead of VGG-19 | patch CNN · MISO-1 EVSP PPM | F-beta measure: 0.321 -> 0.819 | Improved |
| use VGG-Att instead of VGG-19 | patch CNN · MISO-1 FM PPM | F-beta measure: 0.429 -> 0.924 | Improved |
| use VGG-Att instead of VGG-19 | patch CNN · MISO-1 EVP PPM | sensitivity: 0.999 -> 0.999 | No clear effect |
| use VGG-Att saliency instead of VGG-19 saliency | explanation maps · MISO-1 DSP | F-beta measure: 0.000 -> 0.585 | Improved |
| use VGG-Att instead of VGG-19 | explanation maps · MISO-1 EVSP | F-beta measure: 0.000 -> 0.691 | Improved |
| use VGG-Att instead of VGG-19 | explanation maps · MISO-1 GSP | F-beta measure: 0.000 -> 0.350 | Improved |
| use VGG-Att instead of VGG-19 | explanation maps · MISO-2 | ScagDist: 1.000 -> 0.398 | Improved |
| use VGG-Att instead of VGG-19 | patch prediction maps · MISO-2 | sensitivity: 0.557 -> 0.400 | Hurt |
| use VGG-Att instead of VGG-19 | explanation maps · MISO-1 EVP | F-beta measure: 0.475 -> 0.609 | Improved |
Closest papers by meaning
- Streaming Convolutional Neural Networks for End-to-End Learning With Multi-Megapixel Images (TPAMI 2020 · similarity 0.87)
- Efficient Classification of Very Large Images With Tiny Objects (CVPR 2022 · similarity 0.86)
- Processing Megapixel Images with Deep Attention-Sampling Models (ICML 2019 · similarity 0.86)
- I-CEE: Tailoring Explanations of Image Classification Models to User Expertise (AAAI 2024 · similarity 0.83)
- Interpretable Image Classification via Non-parametric Part Prototype Learning (CVPR 2025 · similarity 0.83)
- GRAPE: Graph-Augmented Prototype Explanations for Interactive Medical Image Diagnosis (NeurIPS 2026 · similarity 0.83)
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.
